COMPARISON OF FLAME SPREAD MEASUREMENTS USING THE ASTM E 1321 LIFT AND A REDUCED SCALE ADAPTATION OF THE CONE CALORIMETER APPARATUS BY GEOFFREY MERRYWEATHER SUPERVISED BY MICHAEL SPEARPOINT A thesis submitted in partial fulfilment of the requirements for the Degree of Masters of Engineering in Fire Engineering at the University of Canterbury Department of Civil Engineering University of Canterbury Private Bag 4800 Christchurch, New Zealand Phone: +64 3 364 2250 Fax: +64 3 364 2758 March 2006 iii Abstract A full-scale ASTM E 1321 Lateral ignition and Flame Transport (LIFT) apparatus was constructed and compared with a Reduced scale Ignition and Flame spread Test apparatus (RIFT) adaptation of the cone calorimeter in the vertical position. The objective was to find a low cost and simple alternative to the LIFT apparatus for measuring opposed flow flame spread. Ignition tests were conducted using the LIFT, RIFT and ISO 5657 ignitability apparatus and flame spread experiments were conducted in the LIFT and RIFT. Nine different types of timber based products were tested for ignition and flame spread, and Quintiere?s flame spread model was applied to the results to obtain material properties, such as thermal inertia, flame spread parameter and the minimum heat flux required for flame spread. These materials included plywood, medium density fibreboard (MDF), hardboard, particle board flooring, Melamine (Melteca) covered MDF, New Zealand Rimu, and Beech and New Zealand grown Macrocarpa and Radiata (Monterey) Pine. Further limited tests were conducted on Melteca covered particle board, and a second brand of particle board. The materials in the RIFT were tried with and without preheating to equilibrium. In addition, a view factor for the RIFT was developed, based on earlier work for the cone calorimeter element. The view factor equation was experimentally tested against measured values, and the calculated value was consistently lower than the experimental values, with similar flux profile. The standard procedure is for the material to be preheated before ignition, so that the surface is at equilibrium. The spread of the flame front past points on the sample surface after ignition is recorded, and from the flame front velocity and the model by Quintiere, material specific properties can be derived. The lack of preheating was found to affect the final results, by reducing the flame spread velocity and increasing the scatter in the experimental results. The RIFT gives comparable results to the same materials tested in the LIFT and to the published literature. The results the flame spread parameter and the minimum flux for flame spread are usually higher for the RIFT against the same material in the LIFT. iv There proved to be an effective limit on suitable materials able to be successfully tested in the RIFT to those that have a minimum flux for flame spread of less than 7kW/m 2 , with this limitation is dictated by the flux profile along the sample, and the lower resolution dictated by the smaller size. It is approximately equivalent to a minimum ignition flux of 18kW/m 2 . v Contents Abstract.........................................................................................................................iii Contents .........................................................................................................................v Mathematical symbols ................................................................................................xvi Acknowledgements...................................................................................................xviii 1 Introduction .............................................................................................................1 1.1 The history of flame spread testing................................................................1 1.2 Research objectives........................................................................................2 1.3 Materials chosen ............................................................................................4 2 The theory of flame spread......................................................................................8 2.1 Ignition...........................................................................................................9 2.1.1 Definition of ignition results....................................................................10 2.1.2 Effect of sample orientation on ignition ..................................................13 2.1.3 Effect of sample size................................................................................14 2.1.4 Effect of pilot ignition source ..................................................................15 2.1.5 Effect of moisture content on ignition .....................................................15 2.1.6 Effect of Sample thickness.......................................................................17 2.2 Theories of ignition......................................................................................19 2.2.1 Quintiere, Harkleroad and Walton (1983) and Quintiere and Harkleroad (1984) ..................................................................................................................20 2.2.2 Janssens....................................................................................................24 2.3 Opposed flow (lateral) flame spread............................................................26 2.3.1 Principles of opposed flow flame spread .................................................26 3 The ASTM E1321-97a Lateral ignition and flame transport (LIFT) test apparatus . ...............................................................................................................................32 3.1 History..........................................................................................................32 3.2 Requirements of the ASTM E 1321-97a LIFT apparatus............................32 3.3 The University of Canterbury LIFT apparatus ............................................35 3.3.1 Apparatus description ..............................................................................35 3.3.2 Fuel and air supply...................................................................................37 3.3.3 Flux measurement and calibration ...........................................................40 3.3.4 Heat flux from the pilot flame .................................................................44 3.4 LIFT Operating Procedure...........................................................................44 vi 3.4.1 Ignition test ..............................................................................................44 3.4.2 Analysis of the ignition test .....................................................................47 3.4.3 Flame spread test......................................................................................49 3.4.4 Analysis of flame spread results ..............................................................50 4 The cone calorimeter.............................................................................................52 5 ISO 5657-1987 Ignition test..................................................................................55 6 Reduced Scale ignition and flame test (RIFT) method .........................................57 6.1 Comparison between RIFT and LIFT..........................................................58 6.2 Procedure for using RIFT ............................................................................59 6.2.1 Setup of RIFT ..........................................................................................59 6.2.2 Sample preparation ..................................................................................60 6.2.3 Flux measurement....................................................................................61 6.3 Heat flux and sample angle in the RIFT ......................................................63 6.4 The effect of different element temperatures...............................................63 6.5 The effect of changing the sample angle .....................................................64 6.6 The effect of increasing the separation distance between the sample and element......................................................................................................................66 6.7 Optimum angle of sample in RIFT ..............................................................67 6.8 RIFT test methods........................................................................................69 6.8.1 Ignition test ..............................................................................................69 6.8.2 Flame spread test......................................................................................71 6.9 Calculation of the view factor for the reduced scale LIFT test (RIFT) .......72 6.9.1 Wilson et al, 2002 ....................................................................................72 7 Literature on the LIFT, RIFT and lateral flame spread.........................................77 7.1 LIFT literature..............................................................................................77 7.1.1 Quintiere, J. A simplified theory for generalizing the results from a radiant panel rate of flame spread apparatus. 1981...............................................77 7.1.2 Quintiere, J. Harkleroad, M, Walton D. Measurement of material flame spread properties. 1983 .........................................................................................77 7.1.3 Quintiere, J. Harkleroad, M. New concepts for Measuring flame Spread Properties, 1984.....................................................................................................78 7.1.4 Fowell, A. Interlaboratory test program on ASTM E 1321: Standard test method for measuring material ignition and flame spread properties. Second edition, November 1994........................................................................................78 vii 7.1.5 Dietenberger, M. Experimental and analytical protocol for ignitability of common materials. 1995 .......................................................................................79 7.1.6 Dietenberger, M. Ignitability of siding materials using a modified protocol for LIFT apparatus. 1995 ........................................................................80 7.1.7 Dietenberger, M. Ignitability analysis using the LIFT apparatus and cone calorimeter. 1995...................................................................................................80 7.1.8 Jianmen, Q. Prediction of flame spread test results using the test data from the cone calorimeter. 1990 ...........................................................................81 7.1.9 Goransson U. Using the cone calorimeter to predict flame spread, 199181 7.1.10 Persson G. Predicting lateral flame spread with cone calorimeter, 1993 ..............................................................................................................81 7.1.11 Babrauskas, V. Wetterlund, I. Comparative data from LIFT and cone calorimeter tests for 6 materials, including flame flux measurements. 1999........82 7.1.12 Nisted, T. Flame spread experiments in bench scale; project 5 of the EUREFIC research program. 1991 .......................................................................83 7.2 RIFT literature .............................................................................................84 7.2.1 Azhakesan, A. Shields, T. Silcock, G. Ignition and opposed flow flame spread using a reduced scale attachment to the cone calorimeter. 1998 ...............84 7.2.2 Azhakesan, A. Shields, T. Silcock, G. Combustibility parameters for enclosure lining materials obtained during surface flame spread using reduced scale ignition and flame spread technique. 1998...................................................84 7.2.3 Huynh, VCM. Flame spread measurements of New Zealand timber using an adaptation of the cone calorimeter apparatus. 2003 .........................................84 8 Material test results for manufactured boards .......................................................86 8.1 Medium density fibreboard..........................................................................86 8.1.1 Ignition of MDF.......................................................................................87 8.1.2 Flame spread of MDF ..............................................................................90 8.2 Particle Board (Chipboard)..........................................................................96 8.2.1 Ignition of particle board .........................................................................96 8.2.2 Flame spread for particle board .............................................................102 8.3 Plywood .....................................................................................................108 8.3.1 Ignition of plywood................................................................................108 8.3.2 Flame spread of plywood.......................................................................110 8.4 Hardboard ..................................................................................................115 viii 8.4.1 Ignition of hardboard .............................................................................115 8.4.2 Flame spread of hardboard.....................................................................117 8.4.3 Melteca...................................................................................................121 8.4.4 Ignition of Melteca faced boards ...........................................................122 8.4.5 Flame spread of Melteca........................................................................125 9 Material test results for natural timbers...............................................................131 9.1 New Zealand Beech ...................................................................................131 9.1.1 Ignition of Beech....................................................................................131 9.1.2 Flame spread of Beech...........................................................................133 9.2 Macrocarpa ................................................................................................139 9.2.1 Ignition of Macrocarpa ..........................................................................139 9.2.2 Flame spread of Macrocarpa..................................................................141 9.3 Radiata Pine ...............................................................................................146 9.3.1 Ignition of Radiata Pine .........................................................................146 9.3.2 Flame spread of Radiata Pine ................................................................148 9.4 Rimu...........................................................................................................153 9.4.1 Ignition of Rimu.....................................................................................153 9.4.2 Flame spread of Rimu............................................................................155 10 Discussion of results ......................................................................................159 10.1 Equipment specific issues ..........................................................................159 10.1.1 LIFT ...................................................................................................159 10.1.2 RIFT...................................................................................................161 10.2 Comparison of LIFT, ISO 5657 and RIFT ignition results .......................163 10.2.1 The effect of data reduction of ignition data......................................164 10.2.2 Time to ignition..................................................................................164 10.2.3 Thermal inertia k?c............................................................................168 10.2.4 Minimum ignition flux.......................................................................168 10.2.5 Comparison of correlation values and measured minimum ignition flux " min, . ig q ............................................................................................................169 10.2.6 Apparatus and flame spread...............................................................172 10.2.7 Comparison of ignition parameter ?b?...............................................173 10.2.8 Flame spread parameter ?.................................................................174 ix 10.2.9 Minimum heat flux for flame spread .................................................176 10.3 Material and operational differences .........................................................178 10.3.1 Effect of preheating............................................................................178 10.3.2 Effect of substrates on Melteca and facings on boards......................180 10.3.3 Effect of thickness..............................................................................180 10.4 Proposal for improved apparatus ...............................................................182 11 Comparison of data with published literature................................................184 11.1 Comparison of Ignition results with published literature ..........................185 11.1.1 Thermal inertia (k?c) .........................................................................185 11.2 Comparison of published flame spread results ..........................................190 11.2.1 Flame spread parameter .....................................................................190 12 Conclusions....................................................................................................193 Appendix 1: Procedure for operating University of Canterbury LIFT..................205 Appendix 2: Electrical power and current requirements for an electric LIFT radiant panel ........................................................................................................207 Appendix 3: Results of material tests ....................................................................210 Appendix 4 Flame spread results..........................................................................222 12.1 Summary of flame spread results...............................................................222 12.2 Flame spread data and correlation results..................................................228 Appendix 5 Comparison with published results in the literature ........................357 x List of figures Figure 1: Lateral flame spread .......................................................................................8 Figure 2: Minimum ignition flux .................................................................................11 Figure 3: Critical heat flux...........................................................................................12 Figure 4: Effect of moisture content on ignition time for redwood in LIFT and cone calorimeter. Adapted from Dietenberger (1996) .........................................................16 Figure 5: Heat flux vs. ignition temperature for Radiata Pine. From Moghtaderi et. al (2005)...........................................................................................................................16 Figure 6: Control volume for ignition..........................................................................20 Figure 7: Ignition data plot to find preheating time t * (from Cleary, 1992) ...............23 Figure 8: Opposed flow flame spread..........................................................................27 Figure 9: Main components of LIFT apparatus ...........................................................33 Figure 10: Schematic of LIFT apparatus .....................................................................34 Figure 11: LIFT showing angled gas radiant panel (left) and sample holder (right)...34 Figure 12: Standard ASTM E 1321 LIFT heat flux distribution for calibration .........35 Figure 13: University of Canterbury LIFT apparatus with measuring template in sample holder...............................................................................................................36 Figure 14: UC LIFT burner position adjustment showing rails and leadscrew...........37 Figure 15: Heat flux variation without preloaded air regulator ...................................39 Figure 16: Effect of preloading air regulator to minimum of 8psi on heat flux variation ......................................................................................................................................40 Figure 17: Flux gauge and template.............................................................................41 Figure 18: LIFT calibration profile..............................................................................42 Figure 19: Comparison to UC LIFT calibration flux to ASTM E 1321 standard........43 Figure 20: Variation from standard for calibration of LIFT........................................43 Figure 21: Ignition test of particle board in LIFT........................................................46 Figure 22: Flame spread test........................................................................................49 Figure 23: Flame spread correlation graph ..................................................................51 Figure 24: Cone calorimeter (reproduced from Babrauskas, 2002) ............................53 Figure 25: ISO 5657 ignition apparatus.......................................................................56 Figure 26: Australasian RIFT test, showing location of sample to the cone (reproduced from Huynh, 2003) ..................................................................................57 Figure 27: Heat flux for LIFT vs. RIFT (both at 15 degree angle)..............................58 xi Figure 28: RIFT setup ? plan view ..............................................................................60 Figure 29: RIFT with flux measurement template.......................................................62 Figure 30: RIFT sample holder from rear....................................................................62 Figure 31: RIFT irradiance curve for a sample separation of 45mm, 850?C element temperature and 60? sample angle ...............................................................................63 Figure 32: Effect of changing cone element temperature on received heat flux (70mm separation, 30 degree sample angle) ............................................................................64 Figure 33: Irradiance for a constant 850? element temperature, 70mm separation and changing sample angle.................................................................................................65 Figure 34: Irradiance with constant peak heat flux of 35kW/m 2 and 70mm sample separation .....................................................................................................................66 Figure 35: Change in heat flux with constant element temperature and angle............67 Figure 36: Limitations for flame spread measurement ................................................67 Figure 37: Rift used for ignition testing - rear view ....................................................69 Figure 38: RIFT used for ignition testing ....................................................................70 Figure 39: Flame spread along sample ........................................................................71 Figure 40: Cone calorimeter view factor geometry (from Wilson, 2002) ...................73 Figure 41: RIFT geometry ..........................................................................................74 Figure 42: Comparison of theoretical heat flux ? Equation (38) vs. measured ...........75 Figure 43: Time to ignition for MDF...........................................................................87 Figure 44: Comparison of time to ignition of MDF in ISO 5657 apparatus with Ngu (2002)...........................................................................................................................88 Figure 45: Critical flux for 18mm MDF......................................................................89 Figure 46: Ignition parameters from ASTM e1321-97a for MDF...............................90 Figure 47: LIFT heat flux profile for MDF flame spread test. ....................................91 Figure 48: Flame spread of 18mm MDF in RIFT........................................................92 Figure 49: Flame spread parameter for 18mm MDF in RIFT using ISO ignition data and 40s preheat ............................................................................................................93 Figure 50: Flame spread correlation for 18mm MDF in RIFT and LIFT....................94 Figure 51: Flame spread for 9mm and 18mm MDF in RIFT with low (60s) preheating time ..............................................................................................................................95 Figure 52: Ignition time for 2 brands of particle board in ISO 5657 apparatus...........96 Figure 53: Comparison of ignition times for particle board ........................................97 xii Figure 54: Time to ignition for Pynefloor particle board ............................................98 Figure 55: Time to ignition for Superflake particle board...........................................98 Figure 56: Ignition parameter for Pynefloor particle board........................................99 Figure 57: Ignition parameter for Superflake particle board .......................................99 Figure 58: Critical ignition flux for Pynefloor particle board....................................100 Figure 59: Critical ignition flux for Superflake particle board..................................100 Figure 60: Janssen's critical ignition flux for Pynefloor particle board.....................101 Figure 61: Janssen's critical ignition flux for Superflake particle board ...................102 Figure 62: Comparison of flame spread of 2 brands of particle board ? low preheat time ............................................................................................................................103 Figure 63: Comparison of flame spread for 2 brands of particle board - full preheat time ............................................................................................................................103 Figure 64: Flame spread parameter for Superflake and Pynefloor particle board in RIFT with a 80s preheat time.....................................................................................104 Figure 65: Flame spread correlation for 20mm Particle board in RIFT with full pre- heating time................................................................................................................105 Figure 66: Heat flux profile for LIFT flame spread tests on particle board...............106 Figure 67: Flame spread in LIFT for Pynefloor particle board .................................106 Figure 68: Flame spread correlation for 20mm Pynefloor particle board in LIFT and RIFT...........................................................................................................................107 Figure 69: Time to ignition for 17mm plywood ........................................................108 Figure 70: Critical ignition flux for plywood ............................................................109 Figure 71: Ignition parameter for plywood................................................................110 Figure 72: Flame spread of 17mm plywood in RIFT ................................................111 Figure 73: Comparison of flame spread with different thicknesses of plywood .......112 Figure 74: Flame spread correlation for 17mm plywood in RIFT.............................112 Figure 75: LIFT heat flux profile for plywood and hardboard flame spread tests ....113 Figure 76: Flame spread for 17mm plywood in LIFT ...............................................113 Figure 77: Flame spread correlation for 17mm plywood in RIFT and LIFT ............114 Figure 78: Flame spread correlation for plywood in LIFT........................................114 Figure 79: Time to ignition of hardboard ..................................................................115 Figure 80: Critical heat flux for hardboard................................................................116 Figure 81: Ignition parameters for hardboard............................................................116 Figure 82: Cracking behind the flame front on hardboard ........................................118 xiii Figure 83: Flame spread of hardboard in RIFT ? full preheat time...........................118 Figure 84: Flame spread for hardboard in RIFT ? low preheat time .........................119 Figure 85: Flame spread correlation for hardboard in RIFT......................................119 Figure 86: Flame spread of hardboard in LIFT .........................................................120 Figure 87: Flame spread correlation for hardboard in RIFT and LIFT .....................120 Figure 88: Flame spread behaviour of Melteca faced board......................................122 Figure 89: Time to ignition vs. heat flux of Melteca/MDF board .............................123 Figure 90: Time to ignition for Melteca faced particle board....................................123 Figure 91: Critical ignition flux for Melteca faced MDF ..........................................124 Figure 92: Ignition parameter for Melteca/ MDF board............................................124 Figure 93: Time to ignition vs. heat flux for Melteca faced boards ..........................125 Figure 94: Flame spread of Melteca faced boards in RIFT .......................................126 Figure 95: Comparison of flame spread time to point for MDF vs. particle board, with and without Melteca facing in RIFT..........................................................................127 Figure 96: Flame spread correlation for Melteca faced board in RIFT .....................128 Figure 97: Flame spread correlation for Melteca/ MDF............................................128 Figure 98: Flame spread correlation for Melteca-MDF with a low preheating period ....................................................................................................................................129 Figure 99: LIFT heat flux profile for Melteca faced MDF flame spread test............130 Figure 100: Flame spread correlation for Melteca-MDF in LIFT .............................130 Figure 101: Time to ignition vs. heat flux for NZ Beech ..........................................131 Figure 102: Ignition parameter for Beech..................................................................132 Figure 103: Critical ignition flux for NZ Beech ........................................................133 Figure 104: Flame spread of Beech in RIFT with 55 second preheating period .......134 Figure 105: Flame spread for Beech in RIFT with full preheat.................................135 Figure 106: Flame spread correlation for NZ Beech in RIFT - low preheat time .....135 Figure 107: Heat flux profile for LIFT tests for NZ Beech .......................................136 Figure 108: Flame spread for NZ Beech in LIFT ......................................................137 Figure 109: Flame spread correlation for NZ Beech in LIFT....................................137 Figure 110: Flame spread correlation of Beech in RIFT and LIFT...........................138 Figure 111: Time to ignition for Macrocarpa ............................................................139 Figure 112: Ignition parameter for Macrocarpa.........................................................140 Figure 113: Critical heat flux for Macrocarpa ...........................................................140 Figure 114: Flame spread of Macrocarpa in RIFT with low (40s) preheat time .......141 xiv Figure 115: Flame spread of Macrocarpa in RIFT ? full preheat of 436 seconds .....142 Figure 116: Flame spread correlation for Macrocarpa in RIFT.................................142 Figure 117: Flame spread for Macrocarpa in LIFT ...................................................143 Figure 118: Flame spread correlation for Macrocarpa in LIFT.................................144 Figure 119: Flame spread correlation for Macrocarpa ..............................................145 Figure 120: Time to ignition of Radiata Pine ............................................................146 Figure 121: Critical ignition flux for Radiata Pine ....................................................147 Figure 122: Ignition parameter for Radiata Pine .......................................................147 Figure 123: Flame spread for Radiata Pine in RIFT ? low (40s) preheat..................148 Figure 124: Flame spread for Radiata Pine in RIFT - full preheat ............................149 Figure 125: Comparison of flame spread of Radiata Pine.........................................149 Figure 126: Flame spread correlation for Radiata Pine in RIFT ? low preheat time.150 Figure 127: Flame spread correlation for Radiata Pine in RIFT - full preheat..........150 Figure 128: Flame spread of Radiata Pine in LIFT ...................................................151 Figure 129: Flame spread correlation for Radiata Pine in LIFT................................151 Figure 130: Flame spread correlation for RIFT and LIFT.........................................152 Figure 131: Time to ignition of Rimu........................................................................153 Figure 132: Critical ignition flux for Rimu................................................................154 Figure 133: Ignition parameter of Rimu ....................................................................154 Figure 134: Flame spread along NZ Rimu in RIFT with a low preheat time of 67-80 seconds.......................................................................................................................155 Figure 135: Flux profile of LIFT flame spread test for Rimu and Radiata Pine .......156 Figure 136: Flame spread for Rimu in LIFT .............................................................157 Figure 137: Flame spread correlation for Rimu in RIFT with ISO 5657 ignition ?b? value ? 67-90 seconds preheat time ...........................................................................157 Figure 138: Flame spread correlation for Rimu in LIFT ...........................................158 Figure 139: Marker pins in RIFT...............................................................................163 Figure 140: Comparison of time to ignition for ISO 5657 and LIFT ignition test ....165 Figure 141: Comparison of time to ignition for tests in RIFT and LIFT...................166 Figure 142: Comparison of time to ignition for tests in RIFT and ISO 5657 ignition apparatus ....................................................................................................................167 Figure 143: Comparison of thermal inertia (k?c) in LIFT, ISO 5657 and RIFT.......168 Figure 144: Comparison of minimum ignition flux " min, . ig q ........................................169 xv Figure 145: Comparison of RIFT flame spread correlation and ISO 5657 ignition test results for minimum ignition flux " min, . ig q ...................................................................170 Figure 146: Comparison of LIFT correlation and LIFT ignition test results for minimum ignition flux ...............................................................................................171 Figure 147: RIFT irradiance curve ............................................................................172 Figure 148: Comparison of ignition parameter "b" in ISO 5657 and LIFT...............174 Figure 149: Comparison between flame spread parameter for LIFT and RIFT (using ISO 5657 ign data for RIFT)......................................................................................175 Figure 150: Comparison of flame spread parameter for LIFT and RIFT (using RIFT ignition data) ..............................................................................................................176 Figure 151: Comparison of minimum flux for flame spread " s q in LIFT and RIFT ..177 Figure 152: Minimum ignition flux from ISO 5657 test vs. minimum flux for spread for RIFT .....................................................................................................................178 Figure 153: Effect of preheating on the flame spread correlation for Pynefloor in RIFT ....................................................................................................................................180 Figure 154: Comparison of ignition times for plywood in RIFT test ........................181 Figure 155: Comparison of thermal inertia of plywood ............................................185 Figure 156: Comparison of thermal inertia of particle board ....................................186 Figure 157: Comparison of thermal inertia of MDF..................................................187 Figure 158: Comparison of thermal inertia of hardboard ..........................................188 Figure 159: Comparison of thermal inertia of Radiata (Monterey) Pine...................189 Figure 160: Comparison of thermal inertial of Macrocarpa and NZ Rimu ...............189 Figure 161: Flame spread parameter for plywood.....................................................190 Figure 162: Comparison of flame spread parameter for MDF ..................................191 Figure 163: Flame spread parameter for particle board.............................................192 Figure 164: Regulator and ball valve.........................................................................206 Figure 165: LIFT settings ..........................................................................................206 xvi List of tables Table 1: Materials tested................................................................................................5 Table 2: Pilot flame flux to sample..............................................................................44 Table 3: RIFT setup dimensions ..................................................................................60 Table 4: RIFT polynomial coefficients........................................................................61 Mathematical symbols Symbol Name Units b Ignition parameter s -? Bi Biot number - c Specific heat J/kgK C Flame spread modulus m 3/2 /kWs ? D Thermal penetration depth m Fo Fourier number - h Heat transfer coefficient kW/m?K k Thermal conductivity kW/mK k?c Thermal inertia (kW/m?K)?s " . q Heat flux kW/m 2 t Time s t * Time for thermal equilibrium s T Temperature ?C V Velocity m/s x Distance along sample mm ? Thermal absorptivity - ? Length heated in front of flame m ? Emissivity - ? Flame spread parameter kW 2 /m 3 ? Density kg/m 3 ? Stefan- Boltzmann constant 5.67x10 -8 kW/m?K 4 xvii ? Angle between sample and element ? Subscripts crit Critical ignition value c Convective e Incident on surface f Flame ig At ignition min Minimum value r Radiative peak Peak value s Required for flame spread surf Surface t At t seconds xmm Measured at the x mm position on the sample ? Ambient xviii Acknowledgements There are a large number of people who have made this project and the completion of the Masters possible. In particular, I would like to thank the following people for their assistance and support in helping me to complete this project: ? My parents, Beth and David Merryweather, without their support I could not have done the Masters. ? My wife, Stephanie, who kept the home together by herself while I was away for 2 years. ? Michael Spearpoint, my supervisor, for his help and support over the course of the Masters and particularly this thesis. ? The New Zealand Fire Service Commission who provided a much needed scholarship, and for their ongoing support of the M.E. Fire program. ? The technicians at the University of for their help in various ways and for letting me have use of the workshop for building the LIFT, with particular thanks to the Fire Engineering technicians - Grant and Bob. ? Elizabeth Young and Pauline Anderson, for help, food and coffee. This report is dedicated to my son Robert. 1 1 Introduction 1.1 The history of flame spread testing The study of flame spread along the surface of materials is a key part to fire modelling and prediction of fire behaviour. It forms the underlying basis for many predictions of flashover and room fire growth (Cleary and Quintiere, 1991). It has long been recognised that the rate of flame spread on the surface is an indication of how quickly conditions in a compartment become hazardous. As a result, most building codes require building surfaces to meet minimum flame spread requirements, and the failure to do so has resulted in tragic fires in the past. A well known example is the Stardust Disco fire in Dublin in 1981, which killed 48 people. The wall linings of an alcove were lined with flammable carpet tiles, and the fire rapidly spread once they ignited. (Rasbash, et al 2004). Most of the code requirements are based on an index figure, based on the performance in a test. The formulation of the index, and what is considered to be ?acceptable? performance is largely arbitrary and of little use for further modelling or comparison to other standards. A typical example of this is the ASTM E 84 ?tunnel test? where the flame spread over a surface is given an index value depending on the performance of the material compared with red oak (Janssens, 2005). Such a result allows ranking compared with other materials tested in the same manner, but cannot be easily compared with results from other jurisdictions which have different test methods, and the results cannot also be further developed to give an estimate of the fire behaviour when the material is in use. The wide variety of standards increases the cost and space required in the laboratory, and this is made worse by the fact that many of the tests are similar, but specific to certain materials and situations. There has been a focus on moving to standardised tests, which achieve the objectives of previous test standards, but which can provide data for modelling fire behaviour. This has largely arisen since the development of the ASTM E 1354 cone calorimeter, which has allowed the development of equivalency correlations to previous test methods. The research in this report is specifically on the study of lateral flame spread. This is also known as creeping or opposed flow flame spread, and covers the case where the 2 flame front is moving perpendicularly or against the airflow direction, so that the flame front does not extend over the unburnt material. The airflow is generally buoyancy driven flow, driven by the heat from the fire. The flame spread in this direction is typically in the order of 100 times slower than the case where the flame front is moving in the direction of the fire (Quintiere, 2002) , and hence is less of a concern to life safety in the early stages of fire growth. The protocol which is covered by this research is the Lateral Ignition and Flame Transport (LIFT) test, adopted as the ASTM E 1321 standard. It gives apparent material properties which can be used in modelling fire growth on a surface (Cleary and Quintiere, 1991). These properties include: ? Thermal inertia k?c; ? The minimum heat flux for ignition " min,ig q and the resulting minimum surface temperature for ignition, T ig ? The minimum heat flux for flame spread . " s q and the resulting minimum surface temperature for flame spread, T s ? A material property called the flame spread parameter ? which will allow the calculation of the flame spread rate for a material, if the incident heat flux and material properties are known. 1.2 Research objectives The LIFT apparatus, which forms the basis of the lateral flame spread measurements in the ASTM E 1321-97a standard is bulky, with a footprint of approximately 1.7x0.9m, and requires a gas and compressed air supply for a radiant gas panel.. The numbers of LIFT testing facilities in the world is limited, due to the limited scope of the test. In comparison, the ASTM E 1354 cone calorimeter is the cornerstone of modern fire testing, forming the basis of much of the research into material properties, and hence many of the legislative requirements for materials. Within 10 years of the cone calorimeter being available commercially, nearly 100 were in worldwide service (Babrauskas, 1995). There are at least 3 cone calorimeters in New Zealand, at the University of Canterbury, Building Research Association of New Zealand (BRANZ) 3 and Canesis - formerly the Wool Research Organisation of New Zealand (WRONZ). By comparison, Babrauskas (1995) put the number LIFT testing facilities in the world at 20 in 1995. In addition, the time for ignition testing in the cone calorimeter is significantly less than that of the LIFT, with the test duration approximately half as long (Babrauskas, 1999). The thermocouple control of the cone element also give more consistent performance than is achievable than the gas panel used in the LIFT. Finally, the RIFT and cone calorimeter use smaller sample sizes than the LIFT, allowing more tests to be conducted with a given amount of material. The cone calorimeter has been shown to give good predictions of fire and test performance for materials, electrical cabling, compartment flashover and a wide variety of other parameters. Efforts to apply cone calorimeter data to modelling lateral flame spread have had limited success (Gorranson, 1992). Initial research at FireSERT at the University of Ulster developed the Reduced scale Ignition and Flame spread (RIFT) apparatus, which showed some success. This used the cone calorimeter as the heat source for flame spread testing, in the same manner as the LIFT. Application of the flame spread procedures and theory, developed by Quintiere (1981, 1983, 1984) in the ASTM E 1321 standard gave the flame spread parameter ?, not otherwise obtainable from cone calorimeter data. Further work at the University of Newcastle by Pease (2001) and Huynh (2003) was inconclusive due to the experimental apparatus. In all cases, the RIFT results were compared with data previously published in the literature, and not directly against LIFT tests of the same material. The objectives of this report can be summarised as ? To manufacture and calibrate a full scale LIFT apparatus at the University of Canterbury for the use in this report, and to identify improvements for future research. ? To manufacture and calibrate a RIFT attachment for the cone calorimeter and apply the flame spread theory developed for the LIFT by Quintiere to the results. 4 ? To directly compare the results from the RIFT with the same materials tested in the LIFT, to eliminate a variable in the results. If the research is successful then this will allow laboratories with only the cone calorimeter to derive flame spread properties without new equipment. ? As part of the RIFT research, the development of a view factor for the RIFT, previously developed by Huynh (2003), is to be continued to see if the shortcomings identified in the earlier research can be corrected. The results that will be compared between the two testing methods include: ? The minimum ignition flux min, . " ig q for ignition tests in the LIFT, RIFT and ISO 5657 ignitability apparatus ? The thermal inertia value, k?c from the ignition results ? The minimum flux for flame spread " . s q for LIFT and RIFT tests ? The flame spread parameter, ? Where information in the literature is known, then the values will be compared with previously published data. 1.3 Materials chosen Nine wood based materials were used in the series of tests conducted for this report, with some limited tests conducted on variations of these materials. Most New Zealand manufactured timber based products are Radiata Pine based, as this is the most common species grown in New Zealand. The materials chosen for the tests in this report were based on the availability of previously published data, particularly Huynh (2003) and Azhakesan et al (1998). A secondary objective was to provide some information on New Zealand materials which was otherwise not available. The thickness of the natural timbers used in the RIFT was less than the samples used in the LIFT due to difficulties in inserting the thicker samples into the sample holder. 5 Material Manufacturer and trade name Description Thickness (mm) Density of tested samples (kg/m 3 ) Plywood IPL ?Tuffply? C/D grade untreated radiate Pine 17 487 Particle board (chipboard) Laminex Corp (Fletcher Wood Panels) ?Pynefloor? Laminex Corp (Fletcher Wood Panels) ?Superflake? Radiata Pine based flooring particle board. Radiata Pine based particle board. 20 20 745 673 Medium density fibreboard (MDF) Fletcher Wood Panels ?Customwood? Radiata Pine based standard MDF. 18 620 Melteca faced medium density fibreboard Melteca faced particle board Laminex Corp (Fletcher Wood Panels) Melteca ?Regal? pre - finished shelving White Melamine faced MDF White Melamine pre - finished and clashed shelving from builders merchants 18 18 681 661 Hardboard Unbranded Hardboard fibreboard 5 819 NZ Radiata Pine (Monterey Pine) Clear grade, kiln dried and untreated 16 (RIFT) 20 (LIFT) 425 NZ Macrocarpa Clear grade, 16 (RIFT) 20 (LIFT) 514 NZ Rimu Heart grade 16 (RIFT) 20 (LIFT) 660 NZ Beech 16 (RIFT) 22 (LIFT) 489 Table 1: Materials tested Particle board is made from wood chips held together with a pressure cured adhesive. It is commonly used for flooring and as a substrate for kitset furniture. 6 Two boards of the same nominal type were tested. Both boards are made by the same company ? the Laminex division of Fletcher Wood Panels and are Radiata Pine based, with the Pynefloor having a higher density than the Superflake board used in these tests. According the Material Safety Data Sheets provided for both materials, as the flooring material is expected to be exposed to the weather during construction, a different adhesive is used. The Pynefloor uses a polymerised urea formaldehyde adhesive (up to 15% of the board content) (MSDS Pynefloor 2004), to give weather resistance for up to 8 weeks of exposure, whereas Superflake uses a melamine urea formaldehyde adhesive (up to 13% of the board content (MSDS Superflake, 2004)), and is for internal use only Medium Density Fibreboard (MDF) is a panel product frequently used for cabinetry. It is made from wood fibre, which is exploded with steam before being compressed with a pressure cured adhesive to form a flat, stable wood panel (Youngquist, 1999). The term ?fibreboard?, used in the literature is often poorly defined and the density and properties of fibreboard products can vary greatly depending on the end use and manufacturing process. While ?fibreboard? may refer to MDF, it may also refer to low density fibreboards, of the type commonly used for pinboards and ceiling tiles. As such, the material properties will be different from MDF and will give different test results. Plywood is made from layers of wood, with the grain direction of each ply at 90 degrees to the neighbouring plies, and the layers are joined with a pressure cured adhesive. The grade used for these tests is C/D grade, commonly used for construction. One face is sanded and surface knots are secure or filled. The other face is rough and empty knots are allowed. Plywood can have varying properties, depending on the substrate and surface plies used and the presence of any voids in the material. Hardboard is a compressed fibreboard sheet product, made from wood pulp and fibres, which is exploded with pressure and steam, similar to MDF and this is then compressed using a thermosetting adhesive to form a thin, high density board (Youngquist, 1999), with a characteristic brown colour. It is commonly used for low 7 strength applications such as cupboard backs and drawer bottoms. A common brand name which is often used as a generic name for this type of hardboard is Masonite. Melteca consists of melamine facing on a substrate of either MDF or particle board. It is commonly used for shelving, kitchen cabinets, cupboards and bench tops. Two brands of Melteca faced board were tested, where one uses a MDF substrate, and the other uses particle board. The manufacturer of each board was different so some variation in the performance of the samples due to the differences of the facing material is expected. Macrocarpa (Cypresses Macrocarpa) is a member of the cypress family, and is also known in the US as Monterey cypress. It has the typical cypress smell to it, which is especially noticeable when it burns. The timber is used for boat building, furniture, framing and panelling. It is known for the wild grain and knotty timber it produces, and can be resinous. It is prone to sparking when burning due to the resin pockets. Radiata Pine is a native of California, and is also known there as Monterey Pine. It forms the basis of New Zealand silviculture, making up 95% of the New Zealand timber production in 2004 (anon, MAF statistics, 2005). It forms the basis of almost all New Zealand made framing timber and manufactured wood panels. Rimu is a native New Zealand timber, although other members of the same family are found through the Pacific Rim. It is best known for the dramatic grain of the heart timber and reddish brown colour, although the other Rimu species from the Pacific, principally Fiji and sold in timber suppliers as Pacific Rimu, tends to be plainer and have a lower density. The colour and density varies between the sap wood, and the heart wood. The sap wood and late growth timber tends to be a plain brown and prone to borer infestation. It has been widely used for flooring, panelling and furniture making, and was widely used for framing until the introduction of Radiata Pine from the 1960s. 8 2 The theory of flame spread Flame spread is divided according to the direction of any forced or buoyancy driven flow in regard to the direction of flame travel. In the case of wind assisted or vertical flame spread, the flame extends over the unburnt area, increasing the area affected by the radiation from the flame and hence the preheating of the material. Opposed flow (also called creeping or lateral flame spread) is where the flame is either over the already burnt material or at right angles to it (Figure 1)., so that there is no flame extension over the unpyrolised area The LIFT test, and hence the RIFT test, is concerned with opposed flow flame spread. The theory and principles of opposed flow flame spread is developed later in 2. Figure 1: Lateral flame spread Flame spread is often modelled as a series of piloted ignitions, so that as the area in adjacent to the flame front reaches the critical ignition temperature, then it ignites. This pyrolysis area is heated by gas phase conduction from the flame itself, as well as from the incident heat flux from any external heat source. Therefore, any study of flame spread will involve ignition and ignition theory to at least some degree. Historically, the study of flame spread has resulted in empirical indices, which allow materials to be compared with each other, for that test procedure, but do not lead to material properties suitable for further calculation or modelling. Such tests involving lateral flame spread include BS478 Part 7 and ASTM E 162 (Surface flammability testing of materials using a radiant heat energy source) (Hilado, 1973). These give an index value, based on the extent of flame spread. The ASTM E 1321-97a LIFT test is 9 different in that it provides flame spread theory and the required material properties to allow modelling of flame spread. 2.1 Ignition The study of flame spread and the flame spread rate depends on the ignition properties of the material being tested. The study of ignition can be divided by the means of starting the ignition ? either piloted or unpiloted (autoignition). Most of the study of ignition for fire engineering is piloted ignition, as this is more conservative, as the minimum heat flux required for piloted ignition and the resulting minimum surface temperature is less than the unpiloted equivalents. For laboratory testing, the pilot ignition source is generally a spark, hot wire or a gas flame, depending on the apparatus design. Flame spread is a piloted ignition process. There are a number of models for predicting ignition under varying flux conditions, which are discussed in depth by Babrauskas (2003, 2001), and some of these models have been compared with experimental results by Ngu (2002). The ASTM E 1321-97a LIFT standard is based on the work by Quintiere et al (1981, 1983, 1984) and hence uses the ignition theory developed there. Other theories on ignition have been developed since, which attempt to give more accurate predictions or address the assumptions and simplifications made by Quintiere et al. At the simplest level, a material will ignite when a sufficient quantity of material has pyrolised to reach the lower flammable limit, and in the case of a piloted ignition, a suitable ignition source is available. Using the pyrolysis of the material as an ignition criterion is often not practical, particularly for design purposes. More desirable is a value for the incident heat flux or a temperature at which the item will not ignite, and hence the fire will no longer spread. To achieve this objective, ignition theory can be divided into several areas, depending on the material properties and thickness. Much ignition theory is based on the concept of heat transfer to and from a quasi steady control volume. The final outcome is a function of the material thickness. At one extreme, a thin material can be effectively at constant temperature throughout and so the temperature from the rear face away from an incident heat flux is the same as the front. The resulting temperature rise, and 10 hence time to ignition, can be calculated from the material properties and heat transfer boundary conditions. At the other extreme is a ?thick? material, where there is no heat loss from the back of the sample. The development of the ignition theory Quintiere and Harkleroad (1983) which forms the basis of the LIFT test, given later in this report, is based on ?thermally thick? heat transfer. 2.1.1 Definition of ignition results There are some similar but confusing outcomes from ignition tests, regardless which ignition theory is used. A common requirement is to find the lowest heat flux level at which ignition will occur. Based on this, the likelihood of an item made from that material igniting as a result of the radiant heat flux can be calculated. This forms the basis of much of the legislation limiting heat flux from fires across property boundaries (Babrauskas, 2003) There are 2 ?lowest? heat flux from ignition test data - the minimum heat flux for ignition ( " min, . ig q ) and the critical heat flux ( " . crit q ), and it is important not to confuse the terms. Minimum ignition flux The minimum heat flux for ignition is the lowest level at which the material will ignite within some time limit, and this varies depending on the requirements of the test protocol. It is found by a series of ignition tests, until the sample will no longer ignite within the allowed time limit, and so the plot of the time to ignition vs. the incident heat flux asymptotes to the minimum ignition flux. The value for " min, . ig q is found by the average of the highest level where there is no ignition, and the lowest level that caused ignition within the allowable time, e.g. the ASTM E 1321-97a test standard calls for this value to be found by bracketing to within ?2kW/m 2 . 11 Figure 2: Minimum ignition flux Critical ignition flux Unlike the minimum ignition flux, the critical ignition flux is found by correlation, rather than as an experimental value, and different ignition theories can give different results. It is the theoretical lowest flux which will cause ignition, given an infinitely long time. Implicit is that the material properties do not change significantly during that time. Quintiere?s model assumed a thermally thick material, whereby there is no heat loss out of the material through the back of the sample, and the material behaves like an inert solid with no thermal decomposition or reaction. Plotting the ignition flux against (1/?(time to ignition)) and fitting a straight line to the data ( Figure 3) gives the critical heat flux at the intercept at the x- axis. " min, . ig q Time to ignition (s) Heat flux (kW/m 2 ) 12 Figure 3: Critical heat flux An alternative thermal ignition model which gives a value for the critical heat flux and is also widely used is that of Janssens (1991). It is similar to the Quintiere and Harkleroad model, except the basis for the derivation means that the data is plotted as (incident heat flux) " . e q vs. (time to ignition) 0.55 , rather than (time to ignition) 0.5 as for Quintiere and Harkleroad. The value of the exponent for Janssens theory (0.55) applies if the material is thermally thick. The intercept with the x-axis gives " . crit q as above. As a result, the value for " . crit q is often slightly higher than that given by Quintiere and Harkleroad (1983) for the same data set. (Babrauskas, 2003). Thermal inertia k?c Thermal inertia is defined as the product of the specific heat of the material (c), the thermal conductivity (k) and the density (?). It is a measure of how easily the material absorbs energy and hence how quickly the temperature will rise to ignition. A material with a low thermal inertia will ignite sooner than one with a high thermal inertia under the same conditions and ignition temperature, as more energy must be absorbed to raise the surface temperature by the same amount. (kW/m 2 ) 13 A summary of the theories and research into the ignition of wood is given by Babrauskas (2001, 2003). Seven different ignition theories, including the Quintiere and Harkleroad model used in the LIFT standard and that of Janssens, are compared experimentally by Ngu (2002). 2.1.2 Effect of sample orientation on ignition . The ASTM E 1321 LIFT test uses an ignition test where the sample is held in the vertical plane, whereas the conventional cone calorimeter and ISO 5657 ? 1986 test for ignition of samples subjected to thermal irradiance uses a sample in the horizontal orientation. The detail of each test apparatus is discussed later in this report. It is anticipated that the ignition tests for the RIFT will generally use the horizontal orientation in the ISO 5657 ignition testing apparatus or the cone calorimeter, although ignition tests were conducted in the RIFT in the vertical position, in order to compare the results. Shields, Silcock and Murray (1993) did extensive work on the ISO 5657 ignition apparatus in various orientations, and compared the results to the cone calorimeter. The ISO 5657 apparatus used the ?nodding? gas pilot flame applied at 4 second intervals (Babrauskas, 2003), rather than the spark ignition used in this report, and the time to ignition for the cone calorimeter and the ISO 5657 apparatus was similar, allowing for the margin of error due to the interval of the nodding gas pilot ignition system. Ignition times were similar for horizontal and vertical orientation, for heat fluxes between 50-70kW/m 2 , but at lower flux levels (down to 20kW/m 2 ), there was significant difference with orientation, with vertical samples taking longer to ignite, and inverted samples several times longer still. Babrauskas (2003) notes that the original version of the apparatus developed by the Experimental Building Station (now CSIRO) in Australia had provision for both the gas ignition system and an electric spark ignition system. Dietenberger (1996) found that the time to ignition for samples in the cone calorimeter, with a horizontal sample, was less than that for the LIFT. This is shown in Figure 4, where the results for the ignition in the cone calorimeter are in the horizontal position, and the others are from the LIFT in the vertical position. 14 2.1.3 Effect of sample size The size of the sample can have an effect on the time to ignition, particularly in the vertical orientation. The main difference as the sample height increases is the thickening of the boundary layer and hence the convection coefficient. The main source of heat loss from the sample face is convection, which is proportional to (the height of the sample ) 1/4 . (Babrauskas, 2003) 15 2.1.4 Effect of pilot ignition source The pilot source and type is recognised as a source of error in ignition testing. The error can come from the pilot itself, or the manner in which it is applied, where a directly impinging pilot flame will cause ignition sooner than one which is remote from the sample. Similarly, if a gas flame is used, and the pilot flame is sufficiently large, the heat flux from the pilot flame can lead to a local heat flux which is higher than the nominal value. Similarly, if the pilot source is to far away from the sample, or poorly placed, the ignition time will increase, as the pyrolised material will be diluted with ambient air, decreasing the concentration of flammable gas. In the original research that looked at the flame spread due to radiant panels, such as the LIFT apparatus, Quintiere (1981) used a ?weak? pilot flame. This was formalised experimentally by Quintiere, Harkleroad and Walton (1983) to give the location in the ASTM E 1321 standard. This is mounted above the sample with a backing flange to provide a laminar flow of pyrolised material to the pilot flame. The apparatus can have an additional effect. The original ISO 5657 ignitibility apparatus, as used by Shields, Silcock and Murray (1993) used a pilot gas flame, applied at 10 second intervals to the sample. This automatically introduces an error in the time to ignition, solely due to the resolution of the pilot flame application. The ISO 5657 apparatus used for this report uses a spark ignition, similar to the cone calorimeter, in order to reduce this systematic error. Babrauskas (2003) reports several experiments comparing spark and gas ignition. Gas flame pilots are shown to give more inconsistent results than a spark ignition pilot, with generally shorter ignition times, due to local heating, even with ?non impinging? pilot flames. Babrauskas recommend a spark ignition as the preferred pilot source. 2.1.5 Effect of moisture content on ignition The moisture content of the samples, particularly wood, can have a large effect on the time to ignition. This will also then have an effect on the flame spread rate, since this is also affected by the ignition properties of the material. Dietenberger (1996) conducted experiments with the LIFT and cone calorimeter on redwood, with oven dried (0% MC), 30% and 50% moisture content. The resulting increase in the time to ignition for various moisture content levels is shown in Figure 4. 16 Figure 4: Effect of moisture content on ignition time for redwood in LIFT and cone calorimeter. Adapted from Dietenberger (1996) The effect on moisture content on the ignition temperature (and hence the heat flux for ignition) can be seen in Figure 5 with the ignition temperature increasing with the moisture content. Figure 5: Heat flux vs. ignition temperature for Radiata Pine. From Moghtaderi et. al (2005) Of greater importance to this report, Babrauskas notes that changes in moisture content of timber at room equilibrium do not make a significant difference to ignition 17 time (Babrauskas, 2001) and that differences in ignition times for timber moisture contents between 0-12% MC is lost in the data scatter. 2.1.6 Effect of Sample thickness The thickness of the material affects the time to ignition as it affects the heat loss through the rear of the sample. At one extreme, the material is ?thermally thin?, so that the rear of the sample is effectively at the same temperature as the radiated face. The temperature of the sample then depends on the boundary conditions of both the front and rear face, for example if the face is insulated or under natural or forced air flow. At the other extreme, the material is ?thermally thick?, where the thickness of the material is such that there is no thermal penetration to the rear of the sample and hence no heat loss from the rear face. Some ignition theories for predicting the time to ignition, such as those used to develop the protocol for the ASTM LIFT test require that the material is thermally thick. While materials over 1mm in thickness are generally thermally thick or thermally indeterminate, it depends on the flux level and preheating time. (Babrauskas, 2003). A lower flux level will give a longer time until the ignition temperature is reached, leading to greater thermal penetration. Hence, all materials behave as thermally thick for high heat fluxes with a correspondingly short ignition time, but increasingly behave as thermally thin at lower flux levels and longer ignition times. The time for thermal penetration to depth D is given by Karlsson and Quintiere, (2000) as ?4 2 D t p = (1) Where ? is the thermal diffusivity. c k ? ? = (2) The problem with using Equation (1) to define the minimum thickness is that the properties used to calculate the thermal penetration depth and time are unknown and derived from the ignition test. 18 For wood products, Babrauskas (2003) summarises many of the findings into this area. For wood products, a rule-of-thumb for the thermal depth D is given by Equation (3) with a comment that several exceptions have been found. " . 6.0 500 e q D ? = (3) As the ignition and flame spread theory for the LIFT standard is based on materials being thermally thick, then the thickness of the materials should be chosen to be thermally thick. The ASTM E 1321-97a standard (section X.1.1.1) notes that materials in this category are typically 2-5mm thick, and where the material is less than this, the results apply to the facing and substrate combination as the substrate can have a significant effect on the ignition and flame spread results. 19 2.2 Theories of ignition A large number of ignition theories have been developed, in order to predict the likelihood and time to ignition of materials, with varying degrees of complexity. Two common ones which can give the material properties for use in the flame spread model are outlined in the following sections. The problem of ignition can be treated as a heating problem, with a surface receiving radiation, and losing heat via radiation and convection from the surface (Figure 6), and solving as an energy balance. This also assumes that reactions within the material or on the surface are not significant. This may not always be true if, for example, the production of smoke or water vapour interferes with the received radiation, or alters the material properties. For an inert solid, an approximate solution is possible for the heat transfer to and from the surface, given in equation (4) (Babrauskas, 2003): ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ?=? ? ck th erfc ck th h q TT igig e ig ? ? ? ? 22 . exp1 " (4) Where h is the effective linearised heat transfer coefficient, with both radiative and convection components and ? is the surface absorptivity. The radiation losses are proportional to T 4 and hence a linearised heat transfer coefficient can only be accurate over a short range. 20 Figure 6: Control volume for ignition Babrauskas (2003) notes that the application of Equation (4) is difficult, as there are practical difficulties in experimentally calculating the values. It is mathematically complex, and the Gaussian error function complement (erfc) is prone to rounding error when multiplied with the exponent value. The ignition temperature at ignition is difficult to measure and there is a wide variation in values. Babrauskas (2001) lists a summary of values obtained by various methods, and these range from 228?C to 420?C. The value observed are dependent on the apparatus, orientation, flux level and measurement method (Fangrat et al, 1997). Generally the heat flux is known or more easily measured than the surface temperature at ignition, so any solution should not rely on the ignition temperature for the final result. 2.2.1 Quintiere, Harkleroad and Walton (1983) and Quintiere and Harkleroad (1984) The ignition theory developed by Quintiere, Harkleroad and Walton (1983) and further developed by Quintiere and Harkleroad (1984) forms the basis for the ignition tests in the ASTM E 1321 LIFT ignition tests, and the basis for the material properties Incident heat flux " . e q Heat conducted and radiated away from surface Depth of control volume D 21 used in the flame spread part of the test. It is based on an energy balance into a control volume for an infinite 1 dimensional slab (Figure 6). Some assumptions are made: ? The material under test is modelled as an inert, grey body. ? The temperature in the control volume increases under the received heat flux until it reaches the ignition temperature (T ig )., at which point it ignites ? More generally, the sample is assumed to be ?well behaved? with homogeneous properties, unaffected by the temperature increase to ignition, and the surface does not melt or blister. The material is assumed to ignite once the surface temperature (T surf ) reaches the minimum temperature for piloted ignition (T ig ), and this temperature increase is due to the increase from the external flux and the flame heat flux, i.e.: fesurfig TTTT ?+?= - (5) Taking an energy balance on the control volume, the minimum heat flux to raise the temperature of the surface to the ignition temperature is: )()()( 44 " min, . ??? ???+?= TThTTTThq igsurfsurfcig ?? (6) Where h = linearised heat transfer coefficient, including the radiation component, and the emissivity ? = 1, as the surface chars before ignition. The surface temperature required for ignition can be calculated from Equation (6) if a heat transfer coefficient h is known. Experiments by Quintiere, Harkleroad and Walton (1983) showed the heat transfer coefficient was relatively constant across the temperature range of materials commonly tested in the LIFT at 15 kW/m 2 K. Further work by Dietenberger (1995) showed that in fact the heat transfer coefficient varied along the sample, and that 15 kW/m 2 K was a simplification. As the surface temperature of the sample varies with position, due to the varying heat flux levels, the heat transfer coefficient will also vary as well. This is further complicated by the effect of the gas fired radiant panel in the LIFT, which provides a turbulent flow due to burning gas from the panel (Babrauskas, 2003). 22 Treating the control volume as a semi infinite, thermally thick solid with one dimensional heat flow into the solid gives the Newtonian heating equation in Equation (4), rewritten by Quintiere et al (1983) as Equation (7) - Equation (9) for the surface temperature at ignition, where the absorptivity ? in Equation (4) is assumed to be equal to 1. haterfcatqTTT f xe ige /])exp(1[ " , ?=?=? ? ck h a ? 2 = (7) or htFqTTT f xe ige /)( " . , =?=? ? (8) Where F(t) is the time transient term )()exp(1)( aterfcattF ?= (9) As the time the sample is exposed to the external flux increases, then the time transient term F(t) tends to zero as the material surface approaches equilibrium. From ignition tests (Cleary, 1992) the time transient F(t) is related to the flux levels: ? ? ? == 1 )( " . " min. . tb tF q q e ig * * tt tt ? ? (10) The thermal equilibrium time t * is calculated from the ignition data, where a plot of the ratio of the (minimum heat flux for ignition " min. . ig q / incident heat flux " . e q ) versus (time to ignition) 0.5 (Figure 7). 23 Figure 7: Ignition data plot to find preheating time t * (from Cleary, 1992) As the time t increases then )(tF tends to a value of 1 at long ignition times, shown in Equation (11), hence the plateau in Figure 7. )()exp(1)( aterfcattF ?= ? 1 as ??t (11) As the time t tends to 0 ? i.e. at high heat fluxes with a correspondingly short ignition time, then the exponential term tends to a value of 1 and as a result F(t) tends to a function dependent on the thermal inertia, shown in Equation (12). )()exp(1)( aterfcattF ?= ? ck th ?? 2 as t?0 (12) Hence for the period of time when t is small, then the F(t) can also be written as t ck h tF ?? 2 )( = as t?0 or tbtF =)( (13) Where b is the ignition parameter, given in Equation (14) and this can be used to derive the effective thermal inertia k?c. ck h b ?? 2 = or 2 4 ? ? ? ? ? ? = b h ck ? ? (14) 24 2.2.2 Janssens Although the theory by Quintiere et al is specified for the ASTM E 1321 LIFT standard, some researchers have used the ignition theory by Janssens to derive the required physical properties, a summary of which by Babrauskas (2003) is as follows. The basis is derived from Equation (4) however the absorptivity ? is not necessarily equal to 1, which is assumed in the work by Quintiere. Janssens showed that an approximation for the F(t) term in Equation (9) is 55.0 )(73.01 1 )()exp(1)( ? + ??= at aterfcattF ck h a ? 2 = (15) As the ignition time increases, so that t ig ? ? at the critical ignition flux " . crit q then Equation (15 ?1.0, so then Equation (7) becomes Equation (16) at " . crit q . h q TT crit ig " . ? +? ? (16) And hence the effective heat transfer coefficient can be calculated () ? ? = TT q h ig crit " . .? (17) Rearranging Equation (8) gives Equation (18) for the energy balance on the surface due to radiant ignition. () )(.. " . tFqTTh eig ?=? ? (18) Equation (15) and Equation (17) are substituted into Equation (18) and the common terms, cancel out and Equation (19) gives the time to ignition at a certain heat flux. ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? += 55.0 2 " . " . 73.01 ig crite th ck qq ? (19) 25 As a result of Equation (19), the material properties can be found with the following procedure. Using the time to ignition data for various heat fluxes, as per the Quintiere method, plot t ig -0.55 on the y-axis against " . e q on the x-axis and fit a straight line to the data points. The value for the critical ignition flux " . crit q is the intercept with the x-axis. The ignition temperature T ig is found iteratively from the energy balance at the surface at the minimum ignition flux using Equation (20). )( ( )( 44 " min" . ? ? ? ? ? += ? TT TT h TT q ig ig c ig ? ? (20) The value for the absorptivity ? is specified as ? = 1 in the ASTM LIFT standard. More realistically, wood products are around 0.88, and typical plastics range from 0.64 to 0.92 (Babrauskas, 2003). Given the ignition temperature and absorptivity, then the heat transfer coefficient can be found using Equation (16), and the effective thermal inertia k?c can be calculated from Equation (21). 828.1 " min . 2 73.0 ? ? ? ? ? ? ? ? ? ? = q B hck ig ? where B ig = 1/(slope of data fit line) (21) 26 2.3 Opposed flow (lateral) flame spread 2.3.1 Principles of opposed flow flame spread Work by Quintiere (1981), Quintiere, Harkleroad and Walton (1983) and Quintiere and Harkleroad (1984) led to the ASTM E 1321 LIFT test for lateral flame spread along thick materials burning in still air (i.e., with natural convection rather than forced flow), where flame spread has been modelled as a series of piloted ignitions. The material ignites when the preheated area reaches the ignition temperature (T ig ), and the ignition temperature is assumed to be constant over the temperature range (Quintiere, 2002). The preheating is from the radiation of the flame of already- burning material, any external heat flux and conduction through the material (Delichatsios, 1999), as shown in Figure 8. The heating from the flame is much less than for vertical or wind assisted flame spread, as the flame does not extend over the preheating area, hence there is reduced radiation from the flame to the unpyrolised surface. The main form of heat transfer into the unpyrolised surface is through the gas phase conduction from the flame front (Quintiere, 2002). The heat loss from the area in front of the flame front which has not yet ignited leads to the requirement for a minimum heat flux impinging on the surface and resulting surface temperature for the flame to spread. Without the heat flux to raise the temperature of the material, either from the flame or from an external source, the surface temperature will not be high enough to reach the ignition temperature, and hence there is a limiting flux for the material to give the ignition temperature required for flame spread. This is given in the LIFT/ RIFT test as the heat flux at the point where the flame front ceases to advance. The analysis by Quintiere (1981) involved a moving control volume, shown in Figure 8. 27 Figure 8: Opposed flow flame spread The thickness of the material and any backing substrate can have some effect on the rate of flame spread and behaviour (Quintiere 2002). The ASTM E 1321 LIFT test requires the material to be thermally thick, whereby a thermally thick material does not have significant heat loss through the rear or sides of the sample, so that there is no significant temperature rise on the back of the sample. For thermally thick materials, the flame spread velocity V f at any point along the sample (Equation (28) ) involves an energy balance into the control volume shown ahead of the flame front in Figure 8 (Quintiere et al, 1981) and the energy balance of the control volume can be used to give the material properties, as follows, from Quintiere et al (1983). The solution depends on a number of simplifying assumptions: ? A key assumption is that the length of the control volume is small. This is valid for opposed flow flame spread, as the flame does not extend over the unpyrolised area, so the preheated are is limited by conduction from the flame. The length of the control volume is up to 2mm (Quintiere, 1981). The relatively short length of the control volume (? f in Figure 8) allows the assumption that the incident flux " . e q on the sample is constant over that length, allowing energy absorbed to be calculated and a simple energy balance to be used. Unpyrolised area in front of flame ? length ? f Flame spread direction Incident flux " . e q Flame heat flux " . f q Heat loss from surface ? T ig T Flame D 28 ? The temperature in the control volume increases under the received heat flux until it reaches the ignition temperature (T ig ), at which point it ignites. The flame spread is therefore a series of ignitions along the sample ? the flame front is at the position where the surface temperature is greater than or equal to the ignition temperature. ? The external radiation " . e q varies along the sample, however the radiation from the flame front " . f q is constant and independent of the external radiation. ? More generally, the sample is assumed to be ?well behaved? with homogeneous properties, unaffected by the temperature increase to ignition, and the surface does not melt or blister. The material is assumed to ignite once the surface temperature reaches the minimum temperature for piloted ignition (T ig ), and this temperature increase is due to the increase from the external flux and the flame heat flux: feig TTTT ?+?=? ? (22) The depth of the control volume into the material (D in Figure 8) is equal to the thermal penetration, which is material dependent, and is given in Equation (23) for an exposure of t seconds. The depth of the control volume is ?thin? over the time it takes to heat the length of the control volume from the ambient temperature to the ignition temperature, so the temperature of the control volume is assumed to be the same at any point at a given time. c kt D t ? = (23) If the surface material is at equilibrium with the external heat flux " e q due to a preheating period, then an energy balance on a single unit control volume D.? f in size, moving at the flame front velocity V f is ffigf qTTcDV ?? " . )( ?? ? (24) 29 Taking an energy balance on the control volume, the critical heat flux to raise the temperature of the surface to the ignition temperature in an ?infinite? time is given in Equation (6), reproduced below in Equation (25). )()()( 44 " , . ??? ???+?= TThTTTThq igigigccrite ? (25) Where h = linearised heat transfer coefficient, including the radiation component, and the emissivity ? =1, as the surface chars before ignition. Treating the control volume as a semi infinite, thermally thick solid with one dimensional heat flow into the solid gives the Newtonian heating equation in Equation (7), reproduced below in Equation (26) for the surface temperature up until ignition. haterfcatqTTT f xe surfe /])exp(1[, " . ?=?=? ? ckha ?/ 2 = igsurf TT = at ignition (26) The surface ignites once it reaches the minimum surface temperature for ignition T ig The inverse square root of the flame front velocity is found to be proportional to the difference between the critical ignition flux and the incident flux at that point, as shown in Equation (28). Rearranging Equation (24) and Equation (26) and substituting into Equation (22) gives Equation (27) haterfcatqV ck q TT f xef ff ig /])()exp(1[ " , . 2/1 " . ?+ ? ? ? ? ? ? ? ? ? ? =? ? ? ? ? (27) or the flame front velocity is given as: )](.)([ " , . 2/1 tFqTThCV f xeiigf ??= ? (28) where ff aqC ? " . /1= (29) 30 )()exp(1)( aterfcattF ?= ck h a ? 2 = (30) where C = a material specific constant referred to as the flame heat transfer modulus. The solution to Equation (28) requires the accurate measurement of the preheating area in front of the flame front as it progresses along the sample. This is practically difficult, given the objective is to find the properties of the material to solve Equation (23). As the time the sample is exposed to the external flux increases, then the time transient term F(t) tends to zero as the material surface approaches equilibrium. For this reason, the sample in the LIFT flame spread test is preheated to equilibrium before ignition to simplify Equation (28) The ignition parameter b, calculated from the ignition tests and equation (14), can be used to calculate the required preheating time t * to bring the material to thermal equilibrium, which is given by Equation (31) or from reading from the ignition parameter graph (Figure 7) 2 * 1 ? ? ? ? ? ? = b t (31) As the heat flux along the sample is known, the minimum heat flux required for the flame to spread ( " s q ) is given by the location of the flame front when it self- extinguishes. Similarly, the minimum temperature for flame spread can be calculated from Equation (25), by setting the critical heat flux value in Equation (32) to that of the minimum flux for spread, and then solving for the temperature. )()()( 44 " . ??? ???+?= TThTTTThq ssscs ? (32) 31 From Equation (27) and Equation (28), the flame velocity can be calculated. However it is difficult to do in practice as the length of the pyrolysis area ? f is unknown. However, if the pyrolysis area is assumed to be constant for a given material due to the fact there is no flame impingement on the unpyrolysed area, and the other constants are included in a new value, then equation for flame spread velocity is as below: 2 )( ? ? = TTck V ig f ? ? (33) Where the flame spread parameter ? can be derived from the experimental results using Equation (34). 2 )( 4 Cb ? ? = (34) 32 3 The ASTM E1321-97a Lateral ignition and flame transport (LIFT) test apparatus 3.1 History The ASTM E 1321 standards series originated from the ASTM E 1317 Standard for marine surface finishes and work by Quintiere (1981) and Quintiere and Harkleroad (1982, 1984) and uses the same testing apparatus with the addition of a mathematical model of flame spread developed by Quintiere (1981). The difference between the two standards is principally with the pilot flame location and the addition of a 180mm long flange behind the pilot flame at the top of the sample holder. The ASTM E 1317 standard uses a vertical pilot flame at the hot end of the sample, approximately 10mm away from the sample. The ASTM E 1321-97a standard calls for a horizontal pilot flame, above the top edge of the sample with a vertical backing flange, so that the radiation from the burner flame does not affect the time to ignition results by increasing the heat flux received by the sample. The ASTM E 1317 standard also uses a thermopile in the hood to give basic heat release data, and this is not used in the ASTM E1321-97a test. 3.2 Requirements of the ASTM E 1321-97a LIFT apparatus The standard LIFT testing apparatus uses a gas fuelled diffusion burner and a vertical sample holder, angled at 15 degrees away from each other (Figure 9). The plans for the ASTM E 1317, upon which the ASTM E 1321-97a LIFT test is based, were issued by the National Bureau of Standards and are available from the ASTM as ADJE1317 Adjunct to ASTM1321-97a. The 155 x 800mm sample holder is angled at 15?? 0.25? to the face of the burner (Figure 9). The hot end of the sample is 125mm away from the face of the burner, and offset from the edge of the burner by approximately 125mm (Figure 10). The burner is a flat panel burner 280mm high x 430mm long, with radiant ceramic elements and a reverberant screen. This is mounted so that it allows the burner 33 assembly to rotate to allow horizontal testing, although this is outside the standard test procedure. The rotating ring assembly can be seen behind the burner in Figure 11. The burner offset can be adjusted to give the best match to the standard flux profile (Figure 12). This gives an almost constant flux level for the first 150mm along the sample for use in ignition testing, and then a decreasing flux level to approximately 2% of the peak heat flux. Figure 9: Main components of LIFT apparatus 34 Figure 10: Schematic of LIFT apparatus Figure 11: LIFT showing angled gas radiant panel (left) and sample holder (right) The results from the ignition or flame spread tests can be observed directly or via on observation mirror, which is at the front of the unit and angled to look under the Sample holder Radiant panel Viewing rake Hood Burner rotating ring 35 burner, and this can be seen more clearly in Figure 13. A viewing rake, seen under the sample holder in Figure 11, allows the operator to monitor the flame spread along the face of the sample. The output of the burner can be varied to suit the desired flux levels by altering the air and gas settings. The standard calibration output, measured at the sample face, is shown in Figure 12. Standard LIFT calibration heat flux profile 0 10 20 30 40 50 60 0 50 100 150 200 250 300 350 400 450 500 550 600 650 700 750 800 Distance along sample (mm) H eat fl u x kW / m 2 Figure 12: Standard ASTM E 1321 LIFT heat flux distribution for calibration 3.3 The University of Canterbury LIFT apparatus 3.3.1 Apparatus description The LIFT apparatus used in these experiments was built in the University of Canterbury (UC) over a period of 5 months in 2005. While it follows the design for the ASTM E 1317 apparatus (Adjunct to ASTM E 1317), with the modifications listed for the ASTM E 1321-97a standard, there are some variations. The main variation from the ASTM standard is the fuel used for the burners. The ASTM E 1321-97a LIFT standard calls for methane powered main burner with an air-acetylene pilot flame, whereas the University of Canterbury LIFT uses Liquefied Petroleum Gas (LPG) ? a mixture of propane and butane - for both burners, controlled via mass flow 36 controllers and a variable pressure air regulator (Figure 13). The radiant panel also has adjustable spacing using a leadscrew and hand wheel (Figure 14) to allow for lower heat fluxes for the ignition tests than the LPG fuelled burner can provide. Figure 13: University of Canterbury LIFT apparatus with measuring template in sample holder Observation mirror Mass flow controllers panel and air regulator 37 Figure 14: UC LIFT burner position adjustment showing rails and leadscrew 3.3.2 Fuel and air supply Due to the unavailability of methane, the UC LIFT uses LPG and a compressed air supply for the main radiant panel burner fuel and the pilot flame. The decision to use methane for the burner used in the original research (Quintiere, 1981) and the subsequent ASTM standard was dictated by the wider range of the methane fuelled radiant panel for the ignition test which form part of the experimental procedure. The spacing of the propane fuelled radiant panel used in this case can be adjusted, as the burner can slide on rails and the position is set by using a lead screw (Figure 14) to give the lower flux levels within the limitations of the burner output. This is only for the ignition tests, as the radiant panel can be used with the standard spacing for the flame spread tests. The use of LPG fuel for the burner is justified by the previously published work in the literature. LPG is a mixture of propane and butane, and propane has been used for ignition experiments with the ASTM E1623 ICAL intermediate scale calorimeter (Grand and Mehrafza 2001), which also normally runs on methane or natural gas. The ICAL procedure also uses a fixed sample, and changes the distance from the burner to alter the flux received by the sample. Similarly, in a round robin test of the IMO Resolution A. 653 (16) marine finishes flame spread test (also known as the ASTM E Leadscrew Burner position adjustment hand wheel Standard position stop 38 1317 standard) which uses the same apparatus as the ASTM E 1321 LIFT test, it was noted that one of the laboratories used propane for the radiant panel burner, although this was later changed to methane. (Pauner, 2003). The propane powered radiant panel burner used in the UC LIFT has a lower limit of " 50 . mm q = 25kW/m 2 at the 50mm measuring point when set up in accordance to the standard LIFT spacing (Figure 10), as the burner behaviour becomes unstable and has uneven heating below this level. From the observed conditions during the experiments, it appears that a change to the gas manifold to allow a more even entry of gas to the centre of the gas panel would allow a setting lower than the 25kW/m 2 limit currently imposed, by reducing or eliminating the uneven burning across the face of the panel, giving a lower limit of approximately 20kW/m 2 on LPG. The limitation of " 50 . mm q = 20-25kW/m 2 for the propane fired panel of the UC LIFT is not generally a limitation for the flame spread tests, as the flux level is usually set to 5-10kW/m 2 over the minimum ignition flux, and for timber based products, the minimum ignition flux is usually in the order of 12kW/m 2 (Babrauskas, 2001), and usually higher for ignition tests conducted in the LIFT (Babrauskas, 2003). The level that the panel is set at is not critical, as it is only to increase the resolution by reducing the step size between each measuring point, and to reduce charring during the preheat period. An upper limit was found of " 50 . mm q = 50 kW/m 2 , due to the air supply to the burner. The air compressor used was unable to supply sufficient air for the burner at higher outputs. In trials using an additional air compressor in parallel with the main air supply, the burner was capable of producing heat fluxes greater than " 50 . mm q = 60kW/m 2 , but there was not sufficient air capacity to hold this level for extended periods. At heat flux settings less than " 50 . mm q =30kW/m 2 , initial tests showed erratic output form the burner, where it cycled in a periodic manner, shown in Figure 15. The 39 problem was traced to a combination of the air compressor cycling and variation of the output of the air supply regulator. Changing the air compressor set points to increase the running time, and adding a ball valve to allow some preload of the regulator to a minimum of 8psi gauge pressure significantly reduced this problem. It can be seen in Figure 16 that the heat flux output with the regulator preloaded to 8psi is more stable than without any preload. Measured heat flux at 50mm - no regulator preload 0 5 10 15 20 25 30 35 0 120 240 360 480 600 720 840 960 1080 1200 Time (s) q" 50m m H e a t f l u x ( k W / m 2 ) Figure 15: Heat flux variation without preloaded air regulator 40 Effect of minimum air supply regulator pressure 0 5 10 15 20 25 30 0 60 120 180 240 300 Time (s) q" 50 mm he a t f l ux ( k W / m 2 ) Preload regulator No preload Figure 16: Effect of preloading air regulator to minimum of 8psi on heat flux variation 3.3.3 Flux measurement and calibration Both the LIFT and the RIFT require a measurement of the heat flux at points along the sample face, to allow the calculation of the flame spread velocity and minimum flux level for flame spread. This is done by using a water cooled flux gauge with a template. The template is made from rigid refractory board, with holes drilled at intervals to suit the head of the flux gauge. The head of the flux gauge is to be a snug fit in the holes, without excess clearance; 12.7mm (1/2 inch) diameter holes were used in the templates. The hole spacing for the LIFT template is at 50mm centres, measured from the edge of the flange of the sample holder. The heat flux gauge was bent to allow it to be clamped to the rail behind the sample holder (Figure 17), in order to keep the flux gauge in place and at a consistent angle while measurements were taken. In use, it was found that care had to be exercised to ensure that the head of the heat flux gauge was flush with the face of the template and not shielded by being inside the hole. Other work, outlined in Section 10.1, indicates that having the head of the flux meter proud of the surface of the template, 41 particularly at the cold end of the sample, will improve the accuracy of the measurements. Figure 17: Flux gauge and template For the initial calibration for the radiant panel position, the panel was set to the standard angle and position (Figure 10). With the burner output to 50kW/m 2 , measured at the 50mm position, the flux levels were measured at 50mm intervals along the sample template using a calibrated heat flux meter. When compared to the standard values, the measured values must be within 10%. The burner position can be adjusted by moving the burner frame relative to the sample frame to alter the offset to give the desired distribution and so the sample angle remains at 15? and the gap between the sample (x=0mm) and the face of the burner remains at 125mm. Prior to the flame spread tests, this procedure was repeated for the desired peak heat flux level, so the distribution along the sample was known. The average of 3 calibration runs over 3 days is shown in Figure 18 and Figure 19. The flux profile matches the standard profile well, although the cause of slightly higher value at 350mm is unknown. 42 LIFT calibration 0 10 20 30 40 50 60 0 100 200 300 400 500 600 700 800 Distance along sample (mm) H e a t fl ux (k W / m 2 ) ASTM standard Calibration results Figure 18: LIFT calibration profile From 600mm onwards, the calibration results are outside the 10% limit (Figure 20), however this is beyond the limit of flame spread along the sample for the materials tested, so was not considered to be important. The results of the ASTM E 1317/ IMO resolution A.653 (16) interlaboratory round robin comparison (Pauner, 2003) showed a similar level of variation at the cold end of the sample at the same reference " 50 . mm q heat flux. The tests published by Nisted (1991) and Babrauskas and Wetterlund (1999) also had this issue and adopted the same approach. Due to the low flux levels at the cold end of the sample, the absolute error is less than the percentage variation would indicate. Since the heat flux gauge was held flush with the surface of the measuring template, it records the radiative and a convective component to the total heat flux at the point. Dietenberger (1995c) showed the convection coefficient along the sample varied with position, and hence the convective component diminishes along the sample. Dietenberger recorded a bias up to 13% at the 750mm position for this reason. 43 Comparison of UC LIFT to ASTM E1321-97a standard flux distribution 0 10 20 30 40 50 60 102030405060 Calibration average flux at point (kW/m 2 ) LI FT f l u x a t poi nt ( k W / m 2 ) Figure 19: Comparison to UC LIFT calibration flux to ASTM E 1321 standard Calibration deviation from standard of UC LIFT -5% 0% 5% 10% 15% 20% 25% 0 100 200 300 400 500 600 700 800 Distance along sample V a r i ati o n fr o m stan d a r d val u e (% ) Figure 20: Variation from standard for calibration of LIFT 44 3.3.4 Heat flux from the pilot flame The heat flux from the pilot flame to the sample was measured at the " 50 . mm q position, without the main burner running. The net increase of heat flux to the sample is given in Table 2. The effect of any draft or buoyancy driven updraft was noticeable, with an increase in the net heat flux to the sample of 50%, as the flame moves away from the backing flange. The effect of the pilot burner on the ignition time is within the variation of the experimental error, and variation of the burner output, so cannot be considered a significant source of error. Average flame heat flux 0.14 kW/m 2 Background flux 0.03 kW/m 2 Net heat flux from flame to sample 0.11 kW/m 2 Table 2: Pilot flame flux to sample 3.4 LIFT Operating Procedure The operating procedure for the UC LIFT is given in Appendix 1. The effect of moisture content in the LIFT test is controlled by conditioning the samples at 50?5 % RH and 23?3?C until the product weight stabilises so that the change in mass over a 24hour period is less than 0.1% of the sample mass. All materials used in the experiments conducted in this report were conditioned for at least two weeks, with the conditioning room and laboratory conditions being at the lower end of the allowable range. No oven drying calculations were conducted to get the moisture content of the materials; however Simpson and TenWolde (1999) provide data for the typical moisture content of wood in equilibrium with the surrounding air. For the allowable range given in the ASTM LIFT standard, then the moisture content of the natural timber samples would be between 8.5% and 9.9% MC. The data from Simpson and TenWolde (1999) does not specifically address particle board or Meltica boards, where the high adhesive content may cause some difference to the equilibrium moisture content. 3.4.1 Ignition test The LIFT test comprises of 2 parts, the first is an ignition test where the time to piloted ignition t ig is obtained for various heat fluxes " . e q , and the minimum heat flux 45 for ignition ( " min, . ig q ) is obtained by bracketing the ignition results to find the lowest flux which will cause ignition within 20 minutes. The critical heat flux for ignition ( " . crit q ) is obtained by plotting 1/ ?t ig against " . e q , and the critical heat flux is given by the intercept with the " . e q , axis. The minimum heat flux for ignition ( " min, . ig q ) is given by plotting the time to ignition vs. heat flux and the curve will asymptote at the value for " min, . ig q . Samples, 155 x 155 (+0,-5)mm were conditioned before the tests so that successive weighings, taken 24 hours apart, did not vary by more than 0.1% of the mass. The readings of the temperature and humidity in the conditioning room during the conditioning period indicated that they were generally at the lower end of the allowable range of 23?C ? 3?C and a relative humidity of 50% ?5%. The samples had the edges and back covered in aluminium foil during the test to reduce heat loss, and an insulated backing board is used. (Figure 21) 46 Figure 21: Ignition test of particle board in LIFT The heat flux level at the sample face was measured using a water cooled heat flux gauge and the measuring template, which was fitted to a sample holder frame. The flux measurements for the ignition tests are conducted at the 50mm position ? i.e. 50mm from the edge of the flange on the front face of the sample holder frame. After lighting the pilot burner and main burner, the pilot burner was adjusted to give a flame approximately 180mm long, and after the burner was allowed to equalise for 5 minutes, the flux output was adjusted to give the desired flux level at the 50mm position on the flux measuring template. The equipment was allowed to stabilise to give a constant output prior to testing. The sample was wrapped with foil on the back and sides and placed with an insulated backing board in a second sample holder The measuring template was removed and within 10 seconds the sample was slid into place and timing started. Once the sample ignited, it was removed and extinguished, the burner was adjusted to the next flux level, again using the heat flux gauge and template, and the burner was left to stabilise prior to testing the next sample. 47 A series of tests was done at varying heat flux levels to bracket the minimum flux for ignition to within 2kW/m 2 , and up to the maximum that the burner could deliver. 3.4.2 Analysis of the ignition test The basis for the ignition analysis is the ignition theory of Quintiere et al (1983) given in section 2.2.1. Using the ignition data as above, the time to ignition (t ig ) versus heat flux ( " . e q ) was plotted. The graph asymptoted to the minimum ignition flux ( " min, . ig q ) with a time to ignition limit of 20 minutes. The (minimum ignition heat flux / incident heat flux) ( " min, . ig q ) / ( " . e q ) versus 1/?(t ig ) was plotted and a best fit line through the origin was overlaid on the data. As discussed in the following section, values where ( " min, . ig q )/ ( " . e q )/ >0.8 were ignored as discussed below, if it significantly improved the fit of the data. This was a matter of engineering judgment as to the effect of these outliers due to the long heating times before ignition. The point at which the best-fit line passes ( " min, . ig q )/ ( " . e q ) = 1 is the square root of the preheat time t * for the flame spread test ? i.e. ?( t * ) . The slope of the best fit line was the ignition parameter ?b?, and this was used to solve for k?c, the thermal inertia using equation (14). Reduction of ignition data Since the behaviour of the material changes as the heating time increases, the fit of the data for the correlation to a straight line becomes increasingly less accurate, in particular for the data used in Quintiere et al?s method. The values which best reflect the thermal inertia are those at the higher heat fluxes with the corresponding shorter ignition times. For the most accurate result, some of the data for points close to the minimum ignition flux must be ignored. The problem is that the choice of acceptable 48 data points is not clearly defined, a problem that has occurred for other researchers in the past: ?From the examples shown in Figure x1.2 of the (ASTM E 1321) standard, it is clear that the sloping-line segment should follow only the initial set of data points. Data points for large values of ?t are not to follow the sloping-line segment, but rather should end up fitting the subsequent horizontal line segment. Unfortunately, the standard gives no guidance as to how this is to be done. In our study, we inspected the individual data points and excluded as many from the high end of the time scale as did not follow the same slope as the data points on the left side of the graph? ?. This process is clearly dependent on the judgement of the individual doing the data reduction. This means that, starting with the same raw data; two different operators can produce different values of reduced data.? (Babrauskas and Wetterlund, 1999, page 23) Attempting to define the limit for this report, based on the thermal penetration depth using equation (3) did not give a definite cut-off for acceptable data, especially if there was some scatter. In order to provide consistency, the data was plotted and a regression fitted to the data and the result was visually inspected. As a guide, points above ( " min, . ig q ) / " . e q >0.8 could be discarded if it significantly improved the fit of the data. This value was chosen as it encompassed the values close to the minimum ignition flux where the deviation was the greatest. The R 2 correlation coefficient calculated by the spreadsheet for the fit of the data was used as a guide to what was deemed a ?significant? difference. When considering whether to include a data point, it is again arbitrary, but if the R 2 value improved by more than 0.15 by excluding a data point, then this was deemed to be significant, and the value could be deleted. The difference was generally obvious from visual inspection. The choice of which data points to include can have a significant effect on the ignition parameter ?b? and particularly the preheating time t * . 49 3.4.3 Flame spread test Samples, 800 x 155 (+0,-5) mm were conditioned before the test as above. The samples had the edges and back covered in aluminium foil during the test to reduce heat loss, and an insulated backing board was used. Figure 22: Flame spread test Using a lead pencil, a line was scribed along the centre of the sample, in the long direction, with intervals marked at 25mm spacing, starting at the flange of the sample holder. After igniting the pilot burner and main radiant panel burner, the pilot flame was adjusted to approximately 180mm long, and the radiant panel output to approximately 5-10kW/m 2 over the minimum ignition flux level calculated in the ignition test For the material being tested. The panel was left for 10 minutes to stabilise prior to starting the test. The flux levels at 50mm intervals along the sample were measured, using the flux gauge and measuring template, and then the pilot burner was turned off. The sample holder containing the measuring template or dummy sample was removed and within 10 seconds the sample holder containing the test specimen was slid into 50 place and the timing started. If the sample did not ignite within the preheating time t * calculated from the ignition tests, then the pilot burner was lit or the sample was ignited at the bottom of the hot end using an hand held flame. The time of ignition was recorded and the spread of the flame front was timed along the sample, using the reference points marked on the sample, or by using the viewing rakes and observation mirror. As the flux level is effectively constant over the first 150mm of the sample, the flame spread very rapidly over this section. If it was extensively charred from a long preheating period, then it was difficult to ignite using the pilot burner, in which case, a small pilot flame was applied directly to the hot end of the sample. 3.4.4 Analysis of flame spread results The analysis for the material properties is given in section 2. As noted by Babrasukas (1995), the units used in the flame spread calculations are inconsistent. The sample measuring positions are given in millimetres along the sample, but the calculation of the flame spread parameter value (? ) requires the velocity to be in metres/second. The calculation in this case changed the velocities to metres/second. The velocity of the flame front is calculated using a three point least squares fit calculated from the distance and time results of the nominal point and the points behind and in front ? i.e. x n , x n-1 and x n+1 using Equation (35) to get the average at the point. 3 )( 3 ).( 2 2 t t xt xt V f ? ?? ?? ?? = (35) F(t) was calculated for each interval along the sample, where F(t) = 1 if the time from inserting the sample was greater than the equilibrium time t * . A graph of 1/?V f on the y-axis against ( " e q . F(t)) on the x-axis was plotted and a regression line was fitted to the data (Figure 23) 51 Figure 23: Flame spread correlation graph The lowest value of " . e q .F(t) gave the minimum heat flux for flame spread ( " . s q ), as an alternative to the heat flux at the point of the extent of flame spread on the sample during the test. The point at which the regression line crosses the x-axis gave an alternative means of calculating the minimum ignition flux " min, . ig q . Note that strictly speaking, the value of " min, . ig q was an experimental result, whereas the value from the graph was a correlation, so some difference in the result was expected. * * * * * * * * * * X- intercept Minimum ignition flux from correlation Lowest value is minimum flux required for flame spread " . e q 1/?V f " . e q .F(t) 52 4 The cone calorimeter The cone calorimeter was developed by the National Bureau of Standards (now National Institute of Standards and Technology), part of the US Department of Commerce, in 1982. It was a radical departure from previous heat release measuring systems, and attempted to eliminate some of the problems that previous methods had (Babrauskas, 1993). It was adopted as an ISO standard (ISO5660) in 1992, and forms the basis of an increasing number of legislative tests (Babrauskas, 1992) and models of other standards, such as the ISO 9705 room-corner test (Beyler et al, 1999) and the ASTM E 84 ?tunnel? flame spread test (White and Dietenberger, 2004). Conceptually, it is a simple method, but gives the ability to produce a lot of data with a high degree of automation in the data collection. It has a conical element, which gives the device its name. For conventional fire testing, a sample, generally 100mm*100mm is mounted on a load cell, so that the mass loss during burning can be recorded. The sample is exposed to a heat flux, up to 100kW/m 2 , and the heat release rate, mass loss and toxic gas production is recorded. Further enhancements have included measuring the smoke production, opacity and the effect of atmosphere (vitiated or oxygen enriched) on the burning behaviour. There are a number of key features that make it different from other methods of bench scale fire testing (Figure 24). Most obviously is the use of the conical electric element. Most other devices of this sort at the time used flat panel radiant elements, generally gas fired. Using an electrical element eliminated the need for the fragile refractory ceramic tiles and gas supply, and gave a more consistent control. More importantly, it allowed testing to be done in the horizontal position, particularly for melting materials. The conical shape gave an almost constant heat flux across the face of the sample, yet allowed the smoke and combustion products to escape without interfering with the radiation from the element. The combustion products are collected in a hood, and sampled to measure the oxygen, carbon monoxide and other gas levels. The heat release rate is calculated using oxygen consumption calorimetry, whereby the amount of oxygen consumed by the combustion is measured. The heat released per mol of oxygen consumed is effectively constant at 13.1kJ/kg and hence by measuring the oxygen consumed, the heat release 53 rate for the sample can be calculated. As the sample is mounted on a load cell, the heat release rate can be compared with the mass loss and an effective heat of combustion can be determined. Figure 24: Cone calorimeter (reproduced from Babrauskas, 2002) The pilot ignition is by a spark gap, and the position of this is adjustable to be centred in the middle of a horizontal specimen, or at the top of a vertical specimen. A key advantage of the cone calorimeter against tests such as the LIFT is that the data collection is automatic and extensive, hence more data can be collected, the tests are faster and the results do not require as much judgement or manipulation. The cone calorimeter has successfully been used to model a variety of other tests and standards, giving good results for well behaved materials, such as timber products (Janssens, 2005). Thus far, efforts to model lateral flame spread using only cone calorimeter data (Jianmin, 1990) have generally not been successful (Goransson, 1991, Persson, 1993). 54 While the RIFT as tested here is only concerned with ignition and flame spread, Azhakesan et al (1998) used the RIFT data in conjunction with mass loss readings and calorimetry to give a simultaneous heat release rate as part of a room fire model. 55 5 ISO 5657-1987 Ignition test The ISO 5657 apparatus was originally designed in 1970 in the Experimental Building Services in Australia, now part of the CSIRO (Babrauskas, 2003). The ISO 5657-1987 ignition apparatus is similar to the cone calorimeter, in that it uses a conical electrical element, although the size and shape is slightly different. The element is electronically controlled via a thermocouple controller with a maximum heat flux to the sample of 50kW/m 2 . . The standard ignition method was part of the original British Standard BS 476: 1987: Part 13, from which the ISO 5657-1987 standard is derived. This originally used a small pilot flame, which is dipped above the sample at 4 second intervals ? giving a resolution of ?2 seconds. The ignition in the University of Canterbury ignition apparatus is now a spark, similar to the cone calorimeter, to give a more accurate time for ignition. The pilot flame apparatus, seen on the centre of Figure 25, is still present but was not used for these experiments, or those by Ngu (2002). The test differs from the cone calorimeter in that it uses a weighted platform with a counterweight (seen on the right side of Figure 25) to hold the sample against a flat plate with a circular cut-out, in order to provide a consistent distance from the sample to the element, even if the top of the sample shrinks away. The sample size is also larger, using a 165mm square sample, instead of the 100mm sample used in the cone calorimeter. In both cases, the sample is backed by an insulating board. Only a circular area of 150mm diameter of the sample is exposed the element, with the remainder of the sample shielded by the flat plate. 56 Figure 25: ISO 5657 ignition apparatus The sample is wrapped in aluminium foil. The BS476: 1987 Part 13 standard calls for the top of the sample to be covered with aluminium foil, with a circular cut-out, to prevent the plate from being soiled by residue. A shield is used while the sample is being placed in position, shown on the bench in Figure 25, which is removed and timing started until the sample ignites, or the test duration is over. Replacing the shield extinguishes the burning sample which can then be removed. Conical element. Spark igniter at rear of element ?Nodding? gas pilot flame (not used) Counterweight Temperature controller Removable shield 57 6 Reduced Scale ignition and flame test (RIFT) method The RIFT method is a result of research published by Azhakesan et al (1998) at FireSERT in Ulster, Northern Ireland. The objective was to provide an alternative to the standard LIFT test for spread-of-flame measurements, by using a modified cone calorimeter, rather than requiring the ASTM1321 LIFT test apparatus. The results published from Azhakesan et al (1998) showed some success. Further work was done at the University of Newcastle by Pease (2001) and by Huynh (2003) from the University of Canterbury. These results were inconclusive and showed a lack of resolution in the measurements. The principal problem was identified as the location of the sample, where the hot end of the sample was on the centreline of the cone (Figure 26), so only half the heat flux from the cone was available. The resulting flame spread was in the order of 90mm along the sample (Huynh 2003), and the resolution was insufficient to give accurate data. Figure 26: Australasian RIFT test, showing location of sample to the cone (reproduced from Huynh, 2003) The angle of the sample in the tests in Australia (Huynh 2003, Pease 2001) and Ulster (Azhakesan et al 1998) was at 60 degrees to the cone heater element face, compared 58 with 15 degrees for the conventional LIFT test, which uses a larger flat panel element. The sample size chosen was 100 x 350mm in Ulster, and 100 x 250mm in Australia. 6.1 Comparison between RIFT and LIFT The LIFT uses a sample angle of 15 degrees with a flat radiant gas panel instead of the conical element of the RIFT and the LIFT burner panel size and the sample size are in the order of three times larger than those used in the RIFT. The effect of the smaller scale of the RIFT is shown in Figure 27 where the sample angle and peak heat flux is set at the same values for both sets of apparatus. The heat flux decays more rapidly in the RIFT as the measurement moves away from the element, due to the directional nature and smaller size of the cone and consequently the smaller size of the sample. . Comparing the heat flux of the RIFT with the LIFT shows that the potential peak heat flux is less (the LIFT has a peak up to 50kW/m 2 ) and the heat flux diminishes more quickly with the RIFT. The peak heat flux for the RIFT is largely dictated by the available cone element temperature. A maximum temperature of approximately 850?C was chosen in the interests of a longer element life. Increasing the peak heat flux level does little to increase the effective flame spread length. This is discussed in more detail in Section 6.3. This is due to the directional nature of the cone element. Hea flux distribution along sample for LIFT and RIFT 0 5 10 15 20 25 30 35 40 0 50 100 150 200 250 300 350 400 450 500 550 600 650 700 750 800 Distance along sample (mm) H e a t fl ux (k W / m 2 ) LIFT RIFT Figure 27: Heat flux for LIFT vs. RIFT (both at 15 degree angle) 59 6.2 Procedure for using RIFT The procedure for using the RIFT apparatus is based on the ASTM1321 standard and follows the same methodology. The rear and sides of the samples are wrapped in aluminium foil and a 12mm Kaoboard insulating board is used behind the sample. The sample is inserted, with timing starting once the sample is in place against the end of the sample frame. The retaining screws at the rear of the sample frame are tightened to press the sample against the front flange. 6.2.1 Setup of RIFT The cone calorimeter element was set to the vertical position and the long axis centreline of the RIFT sample holder was inline with the centreline of the cone. The end of the sample closest to the element was in line with the edge of the element, as shown in Figure 28. The angle and sample separation distance x 0 was set by measuring perpendicularly from the face of the cone to the sample, as shown in Figure 28 and Table 3. 60 Figure 28: RIFT setup ? plan view The standard RIFT setup is in Table 3. Angle of sample 60 degrees Separation of sample from face of element (z 0 ) 45mm Peak heat flux (0mm along sample) 35kW/m 2 Typical cone temperature 850?C Cone diameter D cone 160mm Table 3: RIFT setup dimensions The sample holder must be level and parallel to the face of the cone element, when viewed in elevation. 6.2.2 Sample preparation The same protocol was used as required by the ASTM E1321-97a LIFT standard , which calls for samples to be conditioned at 23?C ?3?C and a relative humidity of 50% ?5% so that successive weighings, taken 24 hours apart, do not vary by more than 0.1% of the mass. For the flame spread test, the samples had the back and sides covered in aluminium foil to limit moisture loss from the sample. z 1 = D cone .tan? + z 0 Corner at end of sample and edge of cone are in line z 0 ? D cone 61 6.2.3 Flux measurement The heat flux measurement along the sample was conducted using a template made from fibre cement board with location holes at 25mm centres to take the head of a water cooled flux measuring gauge and this was inserted in into the sample holder (Figure 29 and Figure 30). After turning on the cone element and extraction hood, the cone element was brought up to temperature with 5 minute stabilisation breaks at 400 and 600 degrees, before increasing to the final temperature. The element was running at the final temperature for at least 10 minutes before any readings are taken to allow the temperature of the element and apparatus to stabilise. After checking that there was water flow through the heat flux gauge before taking any heat flux readings, the head of the flux gauge was inserted into the template, ensuring that the face of the heat flux gauge was flush with the face of the template and not inside the hole to ensure accurate readings of the incident flux at each point. The heat flux gauge was left in place for at least 30 seconds to allow an average reading of the heat flux over that time to be taken before moving to the next measuring point. As with the LIFT, and outlined in Section 10.1 having the head of the flux meter proud of the surface of the template at the cold end of the sample will improve the accuracy of the measurements due to the convection boundary layer. The flux readings were compared with the required values from Figure 31 and the cone temperature was altered to give the best match. A polynomial match to the final RIFT heat flux profile gives a 5th Degree Polynomial Fit: y=a+bx+cx 2 +dx 3 ... with the coefficients given in Table 4. a = 34.910273 b = -0.064334216 c = -0.0031889273 d = 2.6422166e-005 e = -8.0288574e-008 f = 8.6335299e-011 Table 4: RIFT polynomial coefficients 62 Figure 29: RIFT with flux measurement template Figure 30: RIFT sample holder from rear 63 Flux profile for RIFT - (60? sample angle, 850?C element, 45mm separation of sample to element 0 5 10 15 20 25 30 35 40 0 50 100 150 200 250 300 350 Distance along sample (mm) H eat f l ux (k W / m 2 ) Figure 31: RIFT irradiance curve for a sample separation of 45mm, 850?C element temperature and 60? sample angle 6.3 Heat flux and sample angle in the RIFT The ASTM E 1321-97a standard test for measuring the spread of flame parameters calls for a peak heat flux 5-10kW/m 2 over the minimum ignition flux ( " min, . ig q ) from the ignition tests. For timber based products the peak heat flux would then be between 18-28 kW/m 2 , given the minimum ignition flux is in the order of 13-18 kW/m 2 (Ngu 2002). Azhakesan et al (1998) used a peak heat flux of 35kW/m 2 , whereas Pease (2000) and Huynh (2003) used a 60kW/m 2 , due to the different experimental setup, which only used half the cone width (Figure 26). The choice in the original tests of 35kW/m 2 was given as the typical heat flux to walls in a fire (Azhakesan et al (1998)). 6.4 The effect of different element temperatures The effect of changing the element temperature as a means of altering the received heat flux on the sample can be seen in Figure 32. The main result is the curve is 64 scaled with a correspondingly lower peak flux value. The point at which the minimum heat flux for flame spread occurs moves slightly towards the hot end of the sample, with little overall difference in the overall length of flame spread, as given in Section 6.7, due to the directional nature of the heat flux profile of the conical element. Effect of varying cone temperature on RIFT heat flux distibution with constant sample separation distance and angle 0 5 10 15 20 25 30 35 40 0 50 100 150 200 250 300 350 Distance along sample (mm) H e a t fl u x (k W / m 2 ) 30 deg sample, 830?C element 30 deg sample, 850?C element Figure 32: Effect of changing cone element temperature on received heat flux (70mm separation, 30 degree sample angle) 6.5 The effect of changing the sample angle The effect of a changing sample angle for a constant cone temperature on the received heat flux is shown in Figure 33. The decreasing angle increases the peak flux level, however does not significantly alter the minimum flux level or the point at which this occurs, due to the directional output of the cone element. 65 Heat flux vs distance along sample for 850 deg cone temperature, 70mm separation 0.0 5.0 10.0 15.0 20.0 25.0 30.0 35.0 40.0 45.0 50.0 0 50 100 150 200 250 300 350 Distance from end of sample (mm) H e a t f l u x (k W/m 2 ) 60 degrees 50 degrees 40 degrees 30 degrees Figure 33: Irradiance for a constant 850? element temperature, 70mm separation and changing sample angle More usefully, when the peak heat flux is kept constant by changing the cone temperature to suit, then the effect can be seen in Figure 34. The directional nature of the cone element can be seen by the location of the peak flux level along the sample, and the terminal heat flux, which remains relatively constant, despite the changing view factor. 66 Measured irradiance for constant peak heat flux, 70mm separation 0 5 10 15 20 25 30 35 40 0 50 100 150 200 250 300 350 Distance along sample (mm) H e a t flu x (k W / m 2 ) Heat flux (15 deg angle) Heat flux (20 deg angle) Heat flux (25 deg angle) Heat flux (30 deg angle) Heat flux (60 deg angle) Figure 34: Irradiance with constant peak heat flux of 35kW/m 2 and 70mm sample separation 6.6 The effect of increasing the separation distance between the sample and element Increasing the distance between the sample and the element should increase the spread of the irradiance along the sample, giving greater resolution by allowing for a greater flame spread, as the minimum flux level for flame spread ( " . s q ) is further along the sample. The directional nature of the cone element shows that this does not occur within the limitations of the desired heat flux level and element temperature. 67 Measured heat flux - 30 deg angle. 850 deg cone temp, 45 & 70mm separation 0 10 20 30 40 50 0 100 200 300 400 Distance along sample (mm) H eat fl ux k W /m 2 45mm separation 70mm separation Figure 35: Change in heat flux with constant element temperature and angle 6.7 Optimum angle of sample in RIFT As the basis of flame spread measurements is to have a series of velocity vs. received heat flux measurements, it follows that the distance along the sample from the point of the peak heat flux (L peak ) to the point where the flame spread stops (L spread ) due to insufficient received flux, should be as long as possible, as shown in Figure 36 Figure 36: Limitations for flame spread measurement 68 The profile of the full scale LIFT test allows the apparatus to be used for ignition testing, as there is an almost flat heat flux curve for the first 150mm of the 800mm sample. This feature is not required in the RIFT, as the cone calorimeter or ISO 5657 ignition apparatus can be used for ignition testing in the conventional manner. The maximum distance along the sample, between the points L peak and L spread is when the peak heat flux coincides with the end of the sample, as the location of L spread remains relatively unchanged due to the directional output of the cone element (Figure 34). There is no advantage to having a plateau in the irradiance profile at the beginning of the sample, as the requirement is for different heat fluxes at each measuring point along the sample; hence the optimum angle chosen between the element face and the sample is therefore 60 degrees, as used by Pease (2001), Huynh (2003) and Azhakesan (1998). 69 6.8 RIFT test methods 6.8.1 Ignition test Previous studies involving the RIFT used either the cone calorimeter in the normal horizontal position (Azhakesan 1998, Huynh, 2003), or the ISO 5657 ignition testing apparatus (Huynh, 2003) to obtain the ignition data. Ignition tests were conducted using the RIFT (Figure 37) to see if it improved the correlation of the data to the LIFT results by having the sample in the same orientation. Figure 37: Rift used for ignition testing - rear view The ignition test used 200 x 95mm samples with the sample holder mounted parallel with the face of the cone element. The spark igniter was level with the top of the exposed face of the sample, and 13mm from the sample face (Figure 38). The ignition test in the cone calorimeter uses samples 100mm x 100mm, with an insulated backing board and holder, and an electric spark pilot, mounted 13mm above the sample face, and the test is generally conducted with the sample mounted horizontally. The 70 ISO 5657 ignition apparatus is similar but uses a 165 x 165mm sample. The sample holder in the RIFT ignition test was set to the centreline of the cone, and the cone element set to give the required heat flux, measured with a template in the centreline of the cone element. The required temperatures matched those previously recorded for this cone element when used for ignition testing in the cone calorimeter. Figure 38: RIFT used for ignition testing The sample was mounted parallel to the cone element, and exposed to a constant heat flux. In the same manner as ignition testing in the LIFT, cone calorimeter and ISO 5657 apparatus, the time to ignition was recorded, and then another flux level was chosen and the procedure repeated, until the sample did not ignite in 20 minutes. The external irradiance ( " . e q ) was plotted against the time to ignition (t ig ) to give the minimum flux for ignition ( " min, . ig q ) where the ignition time is effectively infinite. For thermally thick materials, the x-intercept of " . e q . vs. 1/?t ig gave the critical ignition flux ( " . crit q ) Spark igniter 71 6.8.2 Flame spread test The samples were marked with a line in the centre of the face in the longitudinal direction, and across the sample every 10mm along the sample, to allow the flame spread to be measured during the test. The distance from the end of the sample was marked on the sample every 20mm.so the flame progress along the sample could be monitored. The flux measuring template was removed and sample slid into place, with the timing started from the point at which the sample touches the end of the sample holder and the sample holder screws on the back of the sample holder were then tightened so that the face of the sample is pushed against the front flange. If the sample did not within the preheating time t * , then a 25mm pilot flame was applied to the bottom corner of the hot end of the sample to ignite the sample. The time to ignition and the time taken for the flame front to reach each 10mm gridline, taken at the centreline of the sample was recorded. . Figure 39: Flame spread along sample The test was complete once the flame ceases to advance along the sample and extinguishes, or it reaches the end of the sample. 72 6.9 Calculation of the view factor for the reduced scale LIFT test (RIFT) Some initial investigation into the view factor of the RIFT was done by Huynh (2003), but these were generally unsuccessful, due to the sample being positioned in the centre of the cone element, rather than at one edge. Huynh (2003) suggested that the view factor developed by Wilson et al (2002) for the cone calorimeter could be used to calculate a view factor for the RIFT, and that it may be more successful with the sample mounted fully across the face of the cone element. This has been investigated in the following chapter. 6.9.1 Wilson et al, 2002 Work by Wilson et al (2002) gave the heat flux received by a sample under the cone calorimeter as a function of distance from the cone centre, and the distance between the sample and the face of the cone element, given by equation (36) and equation (37) for the heat flux to a point on the sample. The nomenclature of the geometry is as listed by Wilson (2002), reproduced in Figure 40. a distance from centreline dA 1 Elemental area on the sample?s surface H 2 , H 4 H 2 = z/a H 4 = (h + z)/a h height of the frustum (given as 65 mm for standard cone) q? Local radiant heat flux on the sample surface (W/m 2 ) R 2 , R 4 R 2 = r 2 /a R 4 = r 4 /a r 2 , r 4 radii of the base and top of the frustum (80 mm, and 40 mm, respectively) T Average surface temperature of the heating 73 element (K) Z Distance from the lower base of the element frustum to the sample surface Z 2 , Z 4 Z 2 = 1 + H 2 2 + R 2 2 Z 4 = 1 + H 4 2 + R 4 2 ? Emissivity of the heating element (0.99) ? Stefan-Boltzmann constant, 5.67 ? 10 -8 (W/(m 2 K 4 )) Figure 40: Cone calorimeter view factor geometry (from Wilson, 2002) The heat flux to any point on the sample, offset from the centre of the cone element is given by Equation (36) 4 2 4 2 4 2 4 2 4 2 2 2 2 2 2 2 2 4 1 1 4 1 1 2 1 " T RZ RH RZ RH q ?? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ?+ ?? ? ? ? ? ? ? ? ? ? ?+ ?= (36) The heat flux to a point on the sample centre in the cone calorimeter in normal use is given in equation (37) 4 2 4 2 2 4 2 2 2 2 2 )( " T rhz r rz r q ?? ? ? ? ? ? ? ++ ? + = (37) where the temperature of the cone element is the average temperature of the element in Kelvin. In the case of the RIFT, the separation z will vary along the length of the sample, as the sample is at an angle to the face of the element, shown in Figure 41. 74 Figure 41: RIFT geometry As the sample is at an angle to the element face, the heat flux received by the differential element dA 1 is reduced by the cosine of the angle between the sample and the heating element. The heat flux at a point on the sample is therefore given in Equation (38). )cos(. 4 1 1 4 1 1 2 1 " 4 2 4 2 4 2 4 2 4 2 2 2 2 2 2 2 2 ???T RZ RH RZ RH q ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ?+ ?? ? ? ? ? ? ? ? ? ? ?+ ?= (38) The heat flux on the sample is limited by peak temperature attainable by the element, and this is taken to be 850?C. The distance between the sample and the element (z 0 ), 75 measured perpendicular to the face of the element at the closest edge is used as a measure of the separation of the sample from the element. While Wilson (2002) showed reasonable accuracy for Equation (36) in comparison to measurements taken under the cone calorimeter in the horizontal position, with a sample parallel to the heater face, the relationship becomes increasingly less valid as the sample angle increases. The shape is generally correct, as shown in Figure 42; however the calculated values are significantly lower than the measured values. Comparison between calculated and measured heat flux for the RIFT 0 5 10 15 20 25 30 35 40 45 50 0 50 100 150 200 250 300 350 Distance along sample from end (mm) H e a t fl u x (k W / m 2 ) Measured 30 Calc 30 Measured 40 Calc 40 Measured 50 Calc 50 Measured 60 Calc 60 Figure 42: Comparison of theoretical heat flux ? Equation (38) vs. measured The experimental results by Wilson et al (2002) showed reasonable variation between measured element temperature using thermocouples connected to the element, and an optical pyrometer, which used a wider sampling area than the thermocouples. This showed that there was variation between points around the cone element for surface temperature, and consequently a variation in received flux against the expected value given by Equation (38). Visually, it can be seen from the colour when operating the 76 cone that there is some variation in the element temperature, with the ends of the cone element in the frustum in particular being cooler than the general element temperature. The model is also based on a smooth cone surface as it does not take into account the effect of the spiral element or re-radiation from the supporting elements or the sample holder and takes no account of the element coils. The view factor calculation is not required for the RIFT, which can be set up empirically, based on flux measurements, so no further work was be done in this area. 77 7 Literature on the LIFT, RIFT and lateral flame spread Unlike the cone calorimeter, where a significant number of papers have been published, little has been published on the ASTM 1321 LIFT apparatus. This is largely due to the more limited nature of the test when compared to the cone calorimeter. The cone calorimeter can be used for product comparisons, modelling compartment fire and upholstered furniture behaviour and other legislative tests (Babrauskas, 1992). In comparison, the LIFT is purely a measure of flame spread, and some ignition properties. Some work has been done to model the LIFT using only cone calorimeter data, with mixed success. The publications given here are most specifically dealing with the LIFT and RIFT. A summary of more general ignition and flame spread research is given by Huynh (2003), Ngu (2002) and Babrauskas (2003). 7.1 LIFT literature 7.1.1 Quintiere, J. A simplified theory for generalizing the results from a radiant panel rate of flame spread apparatus. 1981 The basis of the mathematical model for flame spread, which forms the basis of the ASTM E 1321 LIFT and hence the RIFT was initially developed by Quintiere (1981). Although based on the ASTM E 1317 / IMO maritime finishes test apparatus, the derivation makes it generally applicable to any case of opposed flow flame spread under radiant heating, and the resulting equation is that of Equation (28). The requirement for preheating the material in order to aid data reduction is given, but the method of defining the preheating time is given as ?trial and error? by observing the results of the flame spread correlation for linearity. 7.1.2 Quintiere, J. Harkleroad, M, Walton D. Measurement of material flame spread properties. 1983 The work by Quintiere, Harkleroad and Walton in further developing and applying the theories developed by Quintiere (1981) led to the current ASTM LIFT apparatus and procedure. The theory was further developed to be able to calculate the preheating 78 time required using the ignition parameter. The flame spread theory was further developed and applied to particle board as an example material The apparatus was further modified to give more consistent results. Amongst these were the location of the pilot flame and the addition of a flange on the top of the sample holder to improve the consistency of the ignition time and reduce the effect of the pilot flame on the material. 7.1.3 Quintiere, J. Harkleroad, M. New concepts for Measuring flame Spread Properties, 1984 The theories developed earlier by Quintiere ET al (1981, 1983) were applied in a series of experiments to obtain material properties for a wide range of materials. This work formed the basis of the ASTM E 1321 standard. Much of the published material properties in the literature (for example, in Quintiere 2000) come from the results published in this paper. 7.1.4 Fowell, A. Interlaboratory test program on ASTM E 1321: Standard test method for measuring material ignition and flame spread properties. Second edition, November 1994 As part of an ongoing research program into the repeatability of various ASTM test methods, a round robin of tests of 6 materials, held in 4 laboratories was conducted by the ASTM, although the results for one material (particle board composite) were withdrawn as one laboratory had received the wrong material. The materials tested were: ? Douglas Fire plywood ? Fire retardant treated Douglas Fir plywood ? Composite panel ? Type X gypsum wall board ? Rigid polystyrene foam ? Rigid polyurethane foam No detail of the raw experimental data is given, however some statistical analysis was conducted to give some indication of the variability inherent in the ASTM E 1321 test method. Of interest to this research are the results for the flame spread parameter ? , 79 minimum ignition flux and the other properties, which gave an indication of the acceptable range when comparing with other published results in the literature. The results showed large variations in the results for the flame spread parameter, where the highest value was over 2 times the lowest value. The other results showed large variation, but on a smaller scale. 7.1.5 Dietenberger, M. Experimental and analytical protocol for ignitability of common materials. 1995 Dietenberger conducted a number of tests on wood products for the US Forest Service involving the LIFT and cone calorimeter, with the results published in several papers in 1995 and 1996. This report covers a new protocol for calibration of the LIFT, and recommendations for improving the calibration when aligning the burner to match the required flux profile. The use of at high irradiance level was also recommended, to reduce the error due to the time taken to insert the sample into place. The main result was the development of a function for the actual heat transfer coefficient along the material, given in Equation (39). )")(0138.09.13( 50, mme qxh ?= kW/m 2 K, (39) where x is the distance along the sample from the hot end, in metres. The measured convection coefficient component of the heat transfer coefficient is varies along the sample, and the heat transfer coefficient is significantly higher than the standard value of 0.015kW/m 2 K specified in the ASTM E 1321 standard at points closer than 650mm from the hot end of the sample, and it varies significantly with the panel radiant flux. As a result, using equation (6) and the standard heat transfer coefficient value given in the ASTM E 1321 LIFT standard, gives a higher ignition temperature than experimentally measured values. Using the corrected heat transfer coefficient calculated from equation (39) gave ignition temperatures around 70?C less than predicted for the same materials using the ASTM standard value. A theory of ignition was developed that interpolated the results between thermally thick and thermally thin materials, to allow for the effect of thermal penetration on the sample. 80 7.1.6 Dietenberger, M. Ignitability of siding materials using a modified protocol for LIFT apparatus. 1995 The effect of moisture content, apparatus and thickness of the test samples on ignitability were tested for typical US external cladding materials - Cedar, Redwood, Southern Pine, plywood, hardboard and vinyl faced polystyrene foam. A method is proposed to take into account the effect of moisture content on the ignitability of the material, in order to predict the ignition time and thermal properties, given results at an oven dried moisture level. 7.1.7 Dietenberger, M. Ignitability analysis using the LIFT apparatus and cone calorimeter. 1995 The ignition theory for finitely thermally thick materials with radiant heating, convective cooling and an insulated back face was further developed to give a better correlation result. It is a function of the Biot (Bi) and Fourier (Fo) numbers for the material under the test conditions: n thinthicks ig e FFT q q /1. . )""( 1 1 1 " " + +== (40) where ?/)(" . ? ?= TThq igigig (41) )1/()4.068.2( BiBin ++= (42) 2 1 2 4 " ? ? ? ? ? ? = FoBiF thick ? (43) 1 254.01 exp" ?? ? ? ? ? ? + = Bi BiFo F thin (44) 81 The advantage of this theory is that the assumption of thermal thickness is not required, and the thermal thickness can change through the test at different irradiances and still give satisfactory results. The heat transfer coefficient equation was also developed for the cone calorimeter. 7.1.8 Jianmen, Q. Prediction of flame spread test results using the test data from the cone calorimeter. 1990 Jianmen (1990) attempted to model the flame spread on materials using a view factor calculation and material properties derived from cone calorimeter tests. Using the results from the ISO standards LIFT round robin test, the results looked to give a reasonable match, however later work by Goransson (1991) showed poor results in tests of varying materials. The materials tested were chosen to include as wide a range of properties as possible, and the materials were insulating fibreboard, particle board, PVC carpet on gypsum board and transparent PMMA 7.1.9 Goransson U. Using the cone calorimeter to predict flame spread, 1991 Goransson (1991) conducted tests using the LIFT to attempt to verify the flame spread model developed by Jianmen (1990) and had little success. The problems identified included the assumption in the model that the heat transfer from the flame to the unburnt material was via radiation, when gas phase conduction was the main method, and that the choice of values taken from the cone data does not necessarily represent the flame spread accurately. 7.1.10 Persson G. Predicting lateral flame spread with cone calorimeter, 1993 Following the work by Jianmen (1990) and Goransson (1991), Perrson (1993) used a numerical solution to an integral model of surface flame spread, using the experimental data by Goransson. Most of the data required for the model could be obtained from experimental results in the cone calorimeter. These included: ? Heat flux " . e q along the sample ? Preheating time t * 82 ? Material thickness ? Thermal inertia k?c ? Thermal conductivity of the material k ? Ignition temperature T ig ? Flame heat flux parameters called " 0 . f q and k f The flame heat flux parameters are used to define a decaying heat flux from the flame front on the unburnt material, where the heat flux from the flame is given in equation (45). )(exp("" 0 ffff xxkqq ??= (45) A recognised shortcoming was that the flame heat flux parameters are difficult to define experimentally due to the small scale of the heat affected area and a wide combination of " 0f q and k f will give similar results. This is explored further by Delichatsios (1999). The results showed some success in the materials listed for the ASTM E 1317 / IMO marine finishes test, but performed poorly in the LIFT test. While the equipment is similar, apart from the differences in the pilot flame location, the IMO test involves a fixed peak heat flux of 50kW/m 2 and no preheating time before ignition. The preheating was identified as the main reason why the theory did not perform well against the experimental results. The effect of preheating on the material properties was recognised by Babrauskas and Wetterlund (1999). 7.1.11 Babrauskas, V. Wetterlund, I. Comparative data from LIFT and cone calorimeter tests for 6 materials, including flame flux measurements. 1999 A key part of developing the models of opposed flow flame spread is the heat flux from the flame to the surface of the material. Following an earlier literature review into flame fluxes in opposed flow flame spread (Babrauskas, 1995) where little information had been found, particularly for the LIFT, experiments were conducted at 83 SP in Sweden on 6 materials with both the LIFT and cone calorimeter. These materials included ? particle board, ? fire retardant polyurethane foam, ? black PMMA, ? insulating fibreboard, ? cotton fabric with a Kevlar liner over polyurethane foam, ? acrylic pile fabric over polyurethane foam. In addition to measuring the heat flux from the passing flame front to the material, the report covers the requirements of the ASTM standard, and discusses the problems with the standard regarding clarity and definition of some of the terms, and the effect of preheating on the material. Some tests were done without preheating and the data is included, but no calculations were done or conclusions drawn, as it is outside the Standard requirements. The effect of preheating on the material properties was found to affect some of the materials, in particular the foams. It was found that the flame flux is essentially constant between the materials tested, and hence this should not be a major variant in flame spread modelling. 7.1.12 Nisted, T. Flame spread experiments in bench scale; project 5 of the EUREFIC research program. 1991 Nisted conducted a series of LIFT tests, as part of the wider ?European Reaction to Fire Classification? research program, where the 11 materials covered a range of material properties. The report covered only the LIFT tests, the experimental procedure and results obtained. Of particular interest to this report were the tests for ?ordinary? plywood. The raw test data is included by Nisted, which was used to confirm the calculations used in the post experimental processing in this report. 84 7.2 RIFT literature 7.2.1 Azhakesan, A. Shields, T. Silcock, G. Ignition and opposed flow flame spread using a reduced scale attachment to the cone calorimeter. 1998 The paper by Azhakesan et al in 1998 was the first publication on the RIFT method, and all the resulting work stems from it. The method is described, and four materials are tested ? fibreboard, hardboard, plywood and melamine faced particle board. The results are compared with published data for the LIFT and the BS476 part 6 Surface spread of flame test, rather than testing the same material in these tests. The conclusion was that the RIFT produced similar results to the tests and could give a comparative result to the BS476 part 6 test, with the flame spread parameter slightly higher than the LIFT results produced. 7.2.2 Azhakesan, A. Shields, T. Silcock, G. Combustibility parameters for enclosure lining materials obtained during surface flame spread using reduced scale ignition and flame spread technique. 1998 This publication is a continuation of the previous paper, and expands the use of the RIFT. Heat release rate and mass loss rate data was gathered during the flame spread test in the RIFT, allowing the calculation of heat of gasification and the heat release per unit area burnt to be calculated in addition to the opposed flow flame spread properties calculated in the conventional test. 7.2.3 Huynh, VCM. Flame spread measurements of New Zealand timber using an adaptation of the cone calorimeter apparatus. 2003 Some earlier work by Pease (2000) at the University of Newcastle in Australia to replicate the work of the work of Azhakesan et al led to Huynh (2003) using the University of Newcastle apparatus to attempt to derive some properties for New Zealand timber using the RIFT. Huynh used ignition data, which forms part of the flame spread analysis, from Ngu (2002) which used the ISO 5657-1987 ignitability apparatus. Both the work by Pease and Huynh suffered from a shortcoming in the test 85 apparatus, where the sample was exposed to only half the available element area, and consequently the results were inconclusive. A summary of this research is also given by Spearpoint, et al 2005. 86 8 Material test results for manufactured boards This chapter describes the ignition and flame spread results obtained in the tests on manufactured boards with both the LIFT and the RIFT. Ignition tests for each material were conducted at different heat flux in the ISO 5657 apparatus, the LIFT and the RIFT to obtain the material properties required for the flame spread tests. The flame spread tests were conducted in the RIFT and LIFT. A minimum of 4 samples of each material was tested in the RIFT, with 6 samples if allowed by the available material. Three samples of each material were tested for flame spread in the LIFT. Some additional flame spread and RIFT ignition tests were conducted on different thickness materials of the same nominal type, to gauge an indication that the effect of material thickness may have on the results. As particle board is a common material used for research in the literature, two brands of particle board were tested in the RIFT to show the variation within a material and hence some indication of the error when comparing the experimental results to those published in the literature. The effect of preheating the sample on flame spread was investigated by conducting tests in the RIFT with full preheating, given by the ISO 5657 ignition tests, and a low preheating period of a few seconds. The results for manufactured board were generally more consistent with less variation between tests on the same material than those for natural timbers, as expected from their more consistent composition. The results are summarised in Appendix 3 along with the ignition data, and the raw data from the flame spread tests is in Appendix 4 8.1 Medium density fibreboard The MDF used in these tests was 18mm ?Customwood? brand by Carter Holt Harvey and six samples of 18mm ?Customwood? MDF were tested in the RIFT with a low preheat and a full preheat time, and the same material was used in the LIFT to compare results. The ignition results in the LIFT, RIFT and ISO 5657 apparatus used samples of the 18mm MDF. The extent of flame spread along the sample is shown in 87 Figure 48. Some flame spread tests were conducted on 9mm MDF to look at the effect of thickness on the flame spread rate and variation. 8.1.1 Ignition of MDF MDF showed very consistent ignition and flame spread properties when tested, with little variation between runs. The time to ignition for MDF in the LIFT, RIFT and ISO 5657 tests is shown in Figure 43. The results for ignition in the ISO 5657 ignition apparatus are compared with those obtained by Ngu (2002), who also used the ISO 5657 ignition apparatus to test various New Zealand timbers, shown in Figure 44. The ignition results in these tests are within the error limits of the tests by Ngu, showing the material is very consistent. Time to ignition for MDF 0 100 200 300 400 500 600 700 800 10203040506070 Incident flux (kW/m 2 ) Ti m e to i gni ti on t ig (s ) LIFT ISO5657 RIFT Figure 43: Time to ignition for MDF 88 Time to ignition for MDF in ISO5657 compared with Ngu, 2002 0 120 240 360 480 600 720 840 960 1080 1200 10203040506070 Ti m e t o i g ni t i o n t ig (s) Ngu 2002 MDF Figure 44: Comparison of time to ignition of MDF in ISO 5657 apparatus with Ngu (2002) Plotting the incident heat flux ( " . e q ) vs. 1/? (t ig ), shown in Figure 45, gives the critical heat flux ( " . crit q ) from the intercept with the x-axis. 89 Critical heat flux for 18mm MDF 0 0.05 0.1 0.15 0.2 0.25 1020304050607080 1/ ? (t ig ) (s -0 . 5 ) LIFT ISO5657 RIFT LIFT ISO5657 RIFT Figure 45: Critical flux for 18mm MDF Using the protocol from ASTM E 1321-97a, the ignition parameter is calculated (Figure 46) by plotting ( " min, . ig q )/( " . e q ) vs. ?(t ig ). The intercept of a best fit line with ( " min, . ig q )/( " . e q ) = 1 gives the square root of the preheating time ?t * for the time for the sample to reach thermal equilibrium (t * ). The slope is the ignition parameter ?b?. 90 Ignition parameter for MDF in RIFT, LIFT and ISO5657 LIFT = 0.0468x R 2 = 0.9908 RIFT = 0.0529x R 2 = 0.982 ISO = 0.0532x R 2 = 0.9986 0.00 0.20 0.40 0.60 0.80 1.00 1.20 0 5 10 15 20 25 30 ?(t ig ) s 0.5 LIFT ISO5657 RIFT LIFT RIFT ISO5657 Figure 46: Ignition parameters from ASTM e1321-97a for MDF 8.1.2 Flame spread of MDF The LIFT was set to a " 50 . mm q heat flux of 24kW/m 2 and the profile measured using the flux gauge and template, with the resulting profile shown in Figure 47. The RIFT was setup as outlined in section 6.2. 91 Flux distribution for MDF flamespread tests in LIFT 0 5 10 15 20 25 30 0 100 200 300 400 500 600 700 800 Distance along sample (mm) H e a t fl ux (k W / m 2 ) Figure 47: LIFT heat flux profile for MDF flame spread test. The effect of preheating MDF was compared in the RIFT with and without preheating the material fully to thermal equilibrium prior to ignition. The comparison of distance along the sample against time from ignition is shown in Figure 48, where the unpreheated sample has a lower flame spread rate, and the extent of flame spread is less. The lesser extent of flame spread means that the value for " . s q ? the minimum heat flux for flame spread - is higher for the unpreheated sample. It shows that the flame front progressed faster than the material could heat to the minimum ignition temperature, and hence the flame self extinguished before the maximum flame spread distance could be achieved. 92 Flame spread for MDF in RIFT 0 120 240 360 480 600 0 10 20 30 40 50 60 70 80 90 100 110 120 130 140 150 160 170 180 190 200 Distance along sample (mm) Ti m e f r om i gni t i o n ( s ) Full preheat Low preheat Error bars = 2 std diti Figure 48: Flame spread of 18mm MDF in RIFT The effect of preheating can be seen from Figure 49 and Figure 50, where the slope of the data fit line of the flame spread correlation decreases until the material reaches equilibrium, reflecting the slower flame spread rate of the unpreheated material, due to a lower surface temperature. Tests with varying preheating times were not conducted in the LIFT. 93 Flamespread parameter for 18mm MDF in RIFT (ISO ignition data) 40s preheat time Best fit = -19.227x + 281.69 R 2 = 0.7874 0 20 40 60 80 100 120 140 160 0123456789101121314151617 1/ ? V f (m / s ) -0 . 5 q" s Figure 49: Flame spread parameter for 18mm MDF in RIFT using ISO ignition data and 40s preheat Comparison of MDF in RIFT and LIFT The flame spread correlations for the LIFT and RIFT are compared in Figure 50. The RIFT and LIFT produce similar results for both the slope of the data fit line, which leads to the flame spread parameter, and for the values of the minimum flux for flame spread. The RIFT gives a higher value for the minimum ignition flux from the correlation. 94 Flame spread correlation for 18mm MDF with full preheat RIFT y = -4.9425x + 120.06 R 2 = 0.8763 LIFT y = -4.9827x + 107.38 R 2 = 0.8141 0 20 40 60 80 100 120 0 5 10 15 20 25 30 1/ ? V f ( m /s) -0. 5 LIFT RIFT RIFT LIFT Figure 50: Flame spread correlation for 18mm MDF in RIFT and LIFT The effect of sample thickness on flame spread As a check on the effect of sample thickness on the flame spread, the flame spread for 6 pieces each of 9mm and 18mm MDF is compared in the RIFT with no preheating, shown in Figure 51. This shows that for this material, the two thicknesses have similar flame spread results, with the 18mm MDF having a slightly slower flame spread rate than the 9mm MDF. This is expected to be due to thermal penetration of the thinner material, which is backed by a lightweight insulating board, allowing the material to reach a higher temperature at a given time than the thinner material. The 9mm MDF also had a far higher variation between the runs than the thicker material. 95 Flame spread for MDF in RIFT - 18mm vs 9mm with 60s preheating time - 100 200 300 400 500 600 0 20 40 60 80 100 120 140 160 180 200 Distance along sample (mm) Ti m e f r o m i g ni t i o n t ig (s) 9mm MDF 18mm MDF Error bars = 2 std deviations Figure 51: Flame spread for 9mm and 18mm MDF in RIFT with low (60s) preheating time 96 8.2 Particle Board (Chipboard) Samples of 20mm ?Pynefloor? flooring grade particle board were tested in the ISO 5657 ignition testing apparatus, the RIFT and the LIFT. In addition, a different brand (20mm ?Superflake?), of nominally the same material, was also tested in the RIFT and ISO 5657 apparatus to compare the time to ignition and the flame spread rate in the RIFT, and hence the variation between materials of the same nominal type. The results are summarised in Appendix 3 8.2.1 Ignition of particle board As particle board is a commonly used material in the research literature, the two brands of particle board were compared for ignition, to see if there was significant variation between two samples of the same nominal material, and hence gauge the expected variation with published results. The ignition results in the ISO 5657 ignition apparatus is shown in Figure 52 and Figure 53 Time to ignition for particle board in ISO5657 ignition test 0 120 240 360 480 600 720 840 960 1080 1200 1320 10203040506070 Ti m e to i gni ti on t ig (s ) Superflake Pynefloor Figure 52: Ignition time for 2 brands of particle board in ISO 5657 apparatus. 97 Comparison of ignition time for particle board in ISO5657 ignition test 0 120 240 360 480 600 720 840 960 1080 1200 0 120 240 360 480 600 720 840 960 1080 1200 Pynefloor (s) S upe rf l a k e ( s ) Figure 53: Comparison of ignition times for particle board There is a difference between the 2 brands, but it is small and appears to be approximately in proportion to the product density., where the higher density material (Pynefloor) has a longer time to ignition at any given heat flux level. The results of the ignition tests for Pynefloor are shown in Figure 54, where the minimum flux for ignition ( " min, . ig q ) is 13.75kW/m 2 for the ISO 5657 ignition tests. The LIFT and RIFT gave higher values for the minimum ignition flux, in line with the expected findings. The results for Superflake (Figure 55) are similar. 98 Heat flux vs time to ignition for Pynefloor particle board 0 120 240 360 480 600 720 840 960 1080 1200 1320 10203040506070 T i m e t o ig n i t i o n t ig (s) LIFT ISO5657 RIFT Figure 54: Time to ignition for Pynefloor particle board Time to ignition vs heat flux for Superflake particle board 0 120 240 360 480 600 720 840 960 1080 1200 10203040506070 T i m e to i g n i ti o n (s ) ISO5657 RIFT ign Figure 55: Time to ignition for Superflake particle board 99 The ignition parameters for Pynefloor from the different tests are given in Figure 56. The effect of the different ignition tests on the ignition parameter value ?b?, which is the slope of the best fit line can be seen, as well as the preheating time t * , where (t * ) 1/2 is the intercept where the ignition parameter line crosses " . e q / " min, . ig q = 1 . Ignition parameter for Pynefloor particle board LIFT = 0.0509x R 2 = 0.9797 ISO = 0.0364x R 2 = 0.9935 RIFT = 0.0559x R 2 = 0.9769 0.00 0.20 0.40 0.60 0.80 1.00 1.20 0 5 10 15 20 25 30 35 40 ?(t ig ) ( s 0.5 ) LIFT ISO5657 RIFT Figure 56: Ignition parameter for Pynefloor particle board Ignition parameter for Superflake particle board ISO y = 0.0332x R 2 = 0.8569 RIFT y = 0.0507x R 2 = 0.9202 0.00 0.20 0.40 0.60 0.80 1.00 1.20 0 5 10 15 20 25 30 35 ?(t ig ) (s 0.5 ) ISO5657 RIFT ign Figure 57: Ignition parameter for Superflake particle board 100 The critical ignition flux for the two brands of particle board (Figure 58 for Pynefloor, and Figure 59 for Superflake) gives very similar results. Critical ignition flux for Pynefloor particle board ISO y = 0.003x - 0.0104 R 2 = 0.9882 LIFT y = 0.0029x - 0.0047 R 2 = 0.9866 RIFT y = 0.0035x - 0.0152 R 2 = 0.9878 0.00 0.05 0.10 0.15 0.20 0.25 0 10203040506070 Incident flux (kW/m 2 ) ? (t ig ) (s -0 . 5 ) LIFT ISO5657 RIFT Figure 58: Critical ignition flux for Pynefloor particle board Critical ignition flux for Superflake particle board ISO y = 0.0032x - 0.0109 R 2 = 0.9938 RIFT y = 0.0035x - 0.0185 R 2 = 0.9687 0.00 0.05 0.10 0.15 0.20 0.25 0 10203040506070 Incident flux q" e (kW/m 2 ) 1/ ? (t i g ) (s -0 . 5 ) ISO5657 RIFT ign Figure 59: Critical ignition flux for Superflake particle board 101 As a comparison between Janssen?s ignition theory and that of Quintiere, the critical ignition flux " crit q is shown in Figure 58 and Figure 60. Typically, Janssen?s correlation gives a slightly higher value for the critical ignition flux; due to the steeper slope of the data fit line. In this case the values for the Quintiere model and the Janssen?s correlation are 5.5 and 6 kW/m 2 respectively for Pynefloor particle board. The Quintiere model give 3.4kW/m 2 for Superflake, against 5.8kW/m 2 for the Janssen?s? model for the same material. Janssen's critical ignition flux for Pynefloor particle board ISO y = 0.0026x - 0.0151 R 2 = 0.9892 LIFT y = 0.0025x - 0.0112 R 2 = 0.9839 RIFT y = 0.0031x - 0.0229 R 2 = 0.9878 0.00 0.02 0.04 0.06 0.08 0.10 0.12 0.14 0.16 0.18 0 10203040506070 (t ig ) -0 . 5 5 s -0 . 5 5 LIFT ISO RIFT Figure 60: Janssen's critical ignition flux for Pynefloor particle board 102 Janssen's critical ignition flux for Superflake particle board LIFT y = 0.0028x - 0.0162 R 2 = 0.9959 RIFT y = 0.0031x - 0.0247 R 2 = 0.9684 0.00 0.02 0.04 0.06 0.08 0.10 0.12 0.14 0.16 0.18 0 10203040506070 (t ig ) - 0 ..5 5 s -0 . 5 5 ISO RIFT Figure 61: Janssen's critical ignition flux for Superflake particle board 8.2.2 Flame spread for particle board Flame spread in RIFT A preheat time of 80 seconds was used for the low preheat tests. The t * time of 755 seconds from the ISO 5657 ignition was used for the full preheat time for the RIFT tests, and the t * time of 386 seconds from the LIFT ignition test was used for the LIFT flame spread tests. A comparison of the flame spread between the 2 brands of particle board in the RIFT shows a minor difference in the flame spread rate (Figure 62 and Figure 63), which affects the final flame spread properties. This reflects the ignition test results, where the Superflake board had a shorter time to ignition at for any given heat flux. From the flame spread theory, this would give a faster flame spread rate, as shown here. Given that particle board is a common material used in flame spread testing, this shows the limitation in comparing ?equivalent? materials. The effect of preheating the material on the variation of the flame spread rate measured in the RIFT can be seen by comparing Figure 62 and Figure 63. It can also 103 be seen in the amount of data scatter in the flame spread correlation (Figure 64) when compared to the same material which is fully preheated (Figure 65). Flamespread for Superflake and Pynefloor particle board in RIFT - 80s preheat time 0 120 240 360 480 600 720 840 0 20 40 60 80 100 120 140 160 180 Distance along sample (mm) Ti m e f r om i gni t i on ( s ) Superflake Pynefloor Error bars = 2 std deviations Figure 62: Comparison of flame spread of 2 brands of particle board ? low preheat time Flame spread for Pynefloor and Superflake particle board in RIFT - full preheat time 0 60 120 180 240 300 360 0 20 40 60 80 100 120 140 160 180 200 Distance along sample (mm) T i m e f r om i g n i t i on (s ) Superflake Pynefloor Error bars = 2 std deviations Figure 63: Comparison of flame spread for 2 brands of particle board - full preheat time The flame spread correlation for the Pynefloor in the RIFT is affected by outliers at the end of the test (Figure 65) ? i.e. at the period of low flame spread rate. Without 104 these, the results would be a better match to those obtained in the LIFT (Figure 68). The Superflake board was not tested in the LIFT as there was insufficient material to conduct a full suite of LIFT tests. Flame spread parameter for Pynefloor and Superflake particle board in RIFT with low(80s) preheating time Superf lake y = -24.769x + 258.66 R 2 = 0.7712 Pynef loor y = -18.369x + 216.54 R 2 = 0.3647 0 20 40 60 80 100 120 140 160 02468101214 1/ ? V f (s -0. 5 ) Superflake Pynefloor Figure 64: Flame spread parameter for Superflake and Pynefloor particle board in RIFT with a 80s preheat time. 105 Flame spread correlation for 2 brands of particle board in RIFT with full preheating time Pynefloor y = -8.0585x + 153.24 R 2 = 0.63 Superflake y = -7.5808x + 141.11 R 2 = 0.9178 0 20 40 60 80 100 120 140 0 5 10 15 20 25 1/ ? V f ( m /s ) -0. 5 Pynefloor Superflake Pynefloor Superflake Figure 65: Flame spread correlation for 20mm Particle board in RIFT with full pre-heating time. Flame spread for particle board in the LIFT The LIFT was set to the flux profile shown in Figure 66 and the samples preheated for the preheating time based on the LIFT ignition results of t * = 386 seconds prior to ignition. Flux profile for particle board LIFT test 0 5 10 15 20 25 30 0 100 200 300 400 500 600 700 800 Distance along sample (mm) H e a t fl u x (k W / m 2 ) 106 Figure 66: Heat flux profile for LIFT flame spread tests on particle board The resulting flame spread along the sample is shown in Figure 67. Flame spread in LIFT for Pynefloor particle board in LIFT 0 100 200 300 400 500 600 700 800 0 100 200 300 400 500 Position along sample (mm) T i m e f r om i gni t i on ( s ) Test 1 Test 2 Test 3 Figure 67: Flame spread in LIFT for Pynefloor particle board The resulting flame spread correlation for Pynefloor is compared to the same material in the LIFT in Figure 68, where the value of the minimum flux for flame spread is similar, but the RIFT gives a lower correlated value of the minimum ignition flux. In both tests, the material was fully preheated, with the RIFT test using the ISO 5657 ignition test data, and the LIFT using the ignition data from the LIFT ignition tests. 107 Flamespread correlation for Pynefloor particle board LIFT y = -3.6294x + 92.518 R 2 = 0.8742 RIFT y = -8.0585x + 153.24 R 2 = 0.63 0 20 40 60 80 100 120 140 0 5 10 15 20 25 30 35 1/ ? V f ( m /s ) -0 . 5 LIFT RIFT Figure 68: Flame spread correlation for 20mm Pynefloor particle board in LIFT and RIFT 108 8.3 Plywood All samples were from a single sheet of IPL ?Tuffply? 17mm untreated Radiata Pine C/D grade plywood. The smooth (C grade) face was used for the ignition and flame spread tests. 8.3.1 Ignition of plywood The time to ignition for varying flux levels is shown in Figure 69. Time to ignition vs heat flux for plywood in ISO5657, RIFT and LIFT 0 120 240 360 480 600 720 840 960 1080 1200 1320 1440 1560 1680 1800 10203040506070 Ti m e t o i gni t i o n t ig (s ) LIFT ISO5654 RIFT Time limit for ASTMe1321- 97a = 1200s Figure 69: Time to ignition for 17mm plywood As can be seen in Figure 69, the fact that the sample does not ignite within the time limit does not mean that the minimum ignition flux is the lowest point that ignition will occur. In the case of this ignition test, the sample had a glowing spot before finally igniting, and there was considerable charring. The critical ignition flux (Figure 70) for plywood by the 3 ignition methods gave a range of results from 4.6kW/m 2 to 10kW/m 2 , compared with 13.75kW/m 2 for the 109 minimum ignition flux in the ISO 5657 test and 16.25kW/m 2 for the RIFT and LIFT ignition tests. The chart for the ignition parameter (Figure 71) shows less spread of the results than with the particle board (Figure 56), largely due to the minimum ignition flux being similar for all test methods. Values above " min, . ig q / " . e q <0.8 were ignored to improve the data fit. Critical ignition flux for plywood LIFT= 0.0046x - 0.0332 R 2 = 0.9657 RIFT= 0.005x - 0.0493 R 2 = 0.9628 ISO= 0.004x - 0.0181 R 2 = 0.9752 - 0.05 0.10 0.15 0.20 0.25 0.30 0 10203040506070 1/ ? (t ig ) s -0 . 5 LIFT ISO5657 RIFT Figure 70: Critical ignition flux for plywood 110 Ignition parameters for plywood in ISO565, RIFT and LIFT, LIFT = 0.0584x R 2 = 0.924 ISO= 0.0481x R 2 = 0.9594 RIFT = 0.0605x R 2 = 0.8893 0.00 0.20 0.40 0.60 0.80 1.00 1.20 1.40 0 5 10 15 20 25 30 35 40 45 ?(t ig ) s 0.5 LIFT ISO5657 RIFT Figure 71: Ignition parameter for plywood 8.3.2 Flame spread of plywood Flame spread in RIFT The flame spread in the RIFT shows the expected behaviour, with the preheated sample having a faster flame spread rate, due to the higher initial surface temperature. The low preheat time was approximately 65 seconds., compared with the preheat time for the LIFT at 293 seconds, and the full preheat time for the RIFT , based on the ISO 5657 ignition data of 432 seconds. The flame spread along the sample is given in Figure 72, showing the effect of preheating on the flame spread rate and the extent of flame spread. The flame spread rate is sufficiently slow compared with the preheating time that the sample can preheat to equilibrium at the extent of flame spread. The extent of flame spread along the sample is therefore the same in both cases. The velocity of the flame front is less for the sample which is not preheated, reflecting the lower initial surface temperature prior to ignition. 111 Flame spread for 17mm plywood in RIFT 0 120 240 360 480 600 0 50 100 150 200 250 Distance along sample (mm) T i m e fr o m i g n i ti o n (s ) Plyw ood - full preheat Plyw ood - low preheat error bars = 2 std Figure 72: Flame spread of 17mm plywood in RIFT Both 9mm and 17mm plywood were tested without a preheating period, to check if the material thickness had any significant effect on the flame spread. In this case, the difference in the flame spread is within the limits of error and there is no significant difference (Figure 73). Different materials and a long preheating time may alter this result, if the material thickness is less than the thermal penetration depth. 112 Comparison of flame spread of plywood in RIFT 0 120 240 360 480 600 720 840 0 50 100 150 200 250 300 Dista nce a long sa mple (mm) T i m e f r o m i g ni t i on ( s ) 17mm plyw ood 9mm plyw ood Error bars = 2 std deviations Figure 73: Comparison of flame spread with different thicknesses of plywood The resulting flame spread correlation from the RIFT flame spread tests is shown in Figure 74. Flamespread correlation of plywood using ISO ignition data in RIFT y = -10.716x + 137.54 R 2 = 0.799 y = -7.7432x + 127.44 R 2 = 0.8361 0 20 40 60 80 100 120 0 2 4 6 8 101214161820 1/ ? V f ( m /s ) -0 . 5 Ply - full preheat Ply - low preheat Best fit - full preheat Best fit - low preheat Figure 74: Flame spread correlation for 17mm plywood in RIFT 113 Flame spread in LIFT The flame spread tests in the LIFT used the heat flux profile shown in Figure 75 with the resulting flame spread along the sample shown in Figure 76. Heat flux distribution for hardboard and plywood tests in LIFT 0 5 10 15 20 25 0 100 200 300 400 500 600 700 800 Distance along sample (mm) H eat fl u x (k W / m 2 ) Figure 75: LIFT heat flux profile for plywood and hardboard flame spread tests Flame spread for 17mm plywood in LIFT 0 60 120 180 240 300 360 420 480 540 600 660 0 50 100 150 200 250 300 350 400 450 500 550 Distance along sample (mm) Ti m e f r om i g ni t i on ( s ) Test 1 Test 2 Test 3 Figure 76: Flame spread for 17mm plywood in LIFT 114 The flame spread correlation (Figure 77 and Figure 78) shows that the RIFT and LIFT gave similar results, however the RIFT has more data scatter than the LIFT. Flame spread correlation for 17mm plywood in RIFT and LIFT RIFT = -4.727x + 95.641 R 2 = 0.6853 LIFT = -3.6501x + 78.73 R 2 = 0.7502 0 20 40 60 80 100 120 0 5 10 15 20 25 1/ ? V f ( m /s ) -0 . 5 RIFT - full preheat LIFT RIFT LIFT Figure 77: Flame spread correlation for 17mm plywood in RIFT and LIFT Flamespread correlation for plywood in LIFT best fit = -3.6501x + 78.73 R 2 = 0.7502 0 10 20 30 40 50 60 0 5 10 15 20 25 30 1/ ? V ( s -0. 5 ) (correlation) Figure 78: Flame spread correlation for plywood in LIFT 115 8.4 Hardboard The material used here was a generic unbranded material from a builders supply merchant, with a thickness of 5mm. 8.4.1 Ignition of hardboard The sample in the ISO 5657 ignition test was backed with 20mm lightweight Kaoboard low density insulating board. The RIFT ignition test used 12mm lightweight Kaoboard. The LIFT uses a higher density CaSi board, and this may affect any time to ignition comparison, as the material is physically thin in comparison to the other materials tested. Attempts to calculate the thermal thickness of the hardboard showed to much variation to be successful. Quintiere and Harkleroad (1984) found little difference attributable to the density of the backing board ? the expected result for a thermally thick material. Time to ignition vs. heat flux for hardboard in LIFT and ISO5657 0 120 240 360 480 600 720 840 960 1080 1200 1320 1440 1560 10203040506070 T i m e t o ig n i t i o n t ig (s) LIFT ISO5654 RIFT Time limit for ASTMe1321-97a = 1200 s Figure 79: Time to ignition of hardboard 116 Critical heat flux for 5mm hardboard ISO y = 0.0029x + 0.0017 R 2 = 0.9947 LIFT = 0.0028x - 0.006 R 2 = 0.8957 RIFT = 0.0035x - 0.0089 R 2 = 0.9818 - 0.05 0.10 0.15 0.20 0.25 0 10203040506070 1/ ? (t ig ) s -0 . 5 LIFT ISO5657 RIFT Figure 80: Critical heat flux for hardboard The ignition parameter (Figure 81) shows considerable spread between the LIFT and ISO ignition results. The calculated preheat time t * from the ISO 5657 ignition tests of 944 seconds is unrealistically long, as the material chars and collapses before this time can be achieved in the flame spread test. The ignition test conducted in the RIFT gave a preheat time of 977 seconds, compared with the LIFT calculated preheating time of 375s. Ignition parameter for hardboard with LIFT, RIFT and ISO5657 ignition test LIFT = 0.0516x R 2 = 0.919 RIFT = 0.032x R 2 = 0.9737 ISO = 0.0325x R 2 = 0.9787 0 0.2 0.4 0.6 0.8 1 1.2 1.4 0 5 10 15 20 25 30 35 40 ?(t ig ) s 0.5 q" ig .m in /q " e LIFT ISO5657 RIFT Figure 81: Ignition parameters for hardboard 117 8.4.2 Flame spread of hardboard Flame spread of hardboard in the RIFT Five repeat runs were conducted with hardboard in the RIFT, and they produced consistent flame spread results between runs in the RIFT and LIFT (Figure 83 and Figure 86), in spite of the behaviour of the material during preheating and burning. The material cracks behind the flame front area (Figure 82) which allows the pyrolised material to open and potentially allow flames to the rear of the sample. With low preheat times, the area of the flame front appears to be unaffected until after the flame has passed and it does not appear to affect the flame spread correlation (Figure 85). The correlation is less accurate with more scatter when the material is given the full preheat time (Figure 87), possibly due to both the material cracking and increasing thermal penetration of the material, making the assumption that the material is thermally thick less valid. A particular problem is the long preheat time dictated by the RIFT and ISO 5657 ignition results and the calculated ignition parameter (Figure 81) was unobtainable in the RIFT, as the sample charred to such an extent that pieces of the sample fell out and the sample self ignited. The preheating time used was therefore the time calculated from the LIFT ignition tests. 118 Figure 82: Cracking behind the flame front on hardboard The consistent flame spread performance during the initial part of the flame spread test gives a good correlation result (Figure 85), with less data scatter for low preheat times, unlike the other products tested; where the data scatter is greater with a low preheat time. The data scatter in the correlation also increases significantly at the end of the flame spread test, possibly due to thermal penetration, and the material cracking and flames getting behind the sample. Flame spread for 5mm hardboard in RIFT - full preheat 0 120 240 360 480 600 720 840 0 50 100 150 200 250 300 Distance along sample (mm) Ti m e f r om i g ni t i on ( s ) Test 1 Test 2 Test 3 Test 4 Test 5 Figure 83: Flame spread of hardboard in RIFT ? full preheat time Comparing the flame spread for the different preheating times shows the expected behaviour, where the preheated sample has a higher flame spread rate (Figure 84). 119 Flamespread for 5mm hardboard in RIFT - low preheat time 0 120 240 360 480 600 720 840 960 1080 1200 1320 0 50 100 150 200 250 300 Distance along sample (mm) Ti m e f r om i g ni t i on ( s ) Test 1 Test 2 Test 3 Test 4 Test 5 Full preheat Figure 84: Flame spread for hardboard in RIFT ? low preheat time Flamespread correlation for 5mm hardboard in RIFT Low preheat y = -13.346x + 147.53 R 2 = 0.8969 Full preheat y = -5.7127x + 109.41 R 2 = 0.5258 0 20 40 60 80 100 120 0123456789101121314151617181920212 1/ ? V f (s -0 . 5 ) Figure 85: Flame spread correlation for hardboard in RIFT The flame spread correlation for the RIFT with a full preheat period is compared to the results from the LIFT in Figure 87. 120 Flame spread for hardboard in the LIFT The heat flux distribution for the LIFT tests is shown in Figure 75, and the resulting flame spread for hardboard in the RIFT is given in Figure 86. Flame spread for hardboard in LIFT 0 200 400 600 800 1000 1200 0 100 200 300 400 500 600 Distance along sample (mm) Ti m e f r om i g ni t i o n ( s ) Test 1 Test 2 Test 3 Figure 86: Flame spread of hardboard in LIFT The results for the flame spread correlation from the RIFT closely match those for the LIFT (Figure 87), although the LIFT results have much less data scatter. Flamespread correlation for 5mm hardboard in RIFT and LIFT LIFT = -4.8559x + 94.732 R 2 = 0.6981 RIFT = -5.7127x + 109.41 R 2 = 0.5258 0 20 40 60 80 100 120 0 5 10 15 20 25 1/ ? V f (s -0 . 5 ) LIFT RIFT LIFT RIFT Figure 87: Flame spread correlation for hardboard in RIFT and LIFT 121 8.4.3 Melteca The flame spread of two samples of melamine sheet products were compared in the RIFT to determine the variation between materials. One was a generic white Melteca faced board, using a MDF substrate, manufactured by Fletcher Wood Panels, and sold through builder?s merchants and panel suppliers. The other was ?Regal? brand shelving with a particle board substrate, sold as prefinished and clashed boards in hardware and builders supplies. Only the Melteca- MDF sheet was tested in the LIFT, due to the amount of material available from the shelving product. The results from the RIFT tests and the Melteca- MDF results from the LIFT indicated that further testing in the LIFT of the Melteca ? particle board material was not required due to the inconsistent flame spread results. The ignition and flame spread behaviour of Melteca shown in Figure 88 is typical. The facing material starts to bubble at relatively low heat fluxes, insulating the substrate material. The flame front only progresses as the facing peels off or the bubbles split, allowing the pyrolised substrate to escape, and this gives erratic flame spread rates and ignition times. 122 Figure 88: Flame spread behaviour of Melteca faced board 8.4.4 Ignition of Melteca faced boards The addition of the melamine (Melteca) facing affects the ignition and hence the flame spread of the samples. The facing board bubbles and chars at lower flux levels than the underlying material, thus insulating the underlying substrate, however it does not ignite easily. The substrate pyrolizes and as the surface bubbles split, the gas escapes and can ignite. As a result, at low flux levels, the ignition time can become erratic. This behaviour also affects the flame spread, giving erratic flame spread results. The insulating property of the facing material is apparent when comparing the ignition flux to that of the bare substrate material. 123 Time to ignition vs heat flux for Melteca faced MDF 0 120 240 360 480 600 720 840 960 1080 0 10203040506070 Ti m e t o i g ni t i on t ig (s ) LIFT ISO5657 RIFT Figure 89: Time to ignition vs. heat flux of Melteca/MDF board Time to ignition vs. heat flux for Melteca faced particle board 0 100 200 300 400 500 600 10203040506070 T i m e t o i g ni t i on ( s ) ISO5657 RIFT Figure 90: Time to ignition for Melteca faced particle board 124 Critical heat flux for melteca faced MDF LIFT = 0.0018x + 0.0115 R 2 = 0.9841 ISO = 0.0023x - 0.0059 R 2 = 0.9736 RIFT y = 0.0024x + 0.0025 R 2 = 0.9671 0 0.02 0.04 0.06 0.08 0.1 0.12 0.14 0.16 0 5 10 15 20 25 30 35 40 45 50 55 60 65 1/ ? (t ig ) (s -0 . 5 ) LIFT ISO5657 RIFT Figure 91: Critical ignition flux for Melteca faced MDF The results are given in Appendix 3. Ignition parameter for LIFT, RIFT and ISO5657 for Melteca/ MDF ISO = 0.0429x R 2 = 0.9806 LIFT = 0.0423x R 2 = 0.9171 RIFT = 0.0512x R 2 = 0.9921 0.00 0.20 0.40 0.60 0.80 1.00 1.20 0 5 10 15 20 25 30 35 ?(t ig ) s 0.5 LIFT ISO5657 RIFT Figure 92: Ignition parameter for Melteca/ MDF board 125 Comparison of ignition results between two substrate materials Comparing the time to ignition of the two Melteca boards with the different substrate materials (Figure 93) shows that the particle board based material ignited faster than the MDF based board, and this is reflected in the flame spread rate, as discussed in the following section. Comparison of time to ignition vs heat flux for Melteca faced boards in ISO5657 test 0 120 240 360 480 600 720 840 960 1080 0 10203040506070 Ti m e t o i g ni t i on ( s ) P/ board substrate MDF substrate Figure 93: Time to ignition vs. heat flux for Melteca faced boards 8.4.5 Flame spread of Melteca Flame spread in RIFT Six samples of each material were tested in the RIFT, both with a preheating period and with no preheating prior to ignition. It was not possible to derive a flame spread correlation for Melteca faced boards in the RIFT with no preheating, as there was no linearity and it was not possible to fit a line to the data. The effect of the substrate can be seen in Figure 94, where the MDF based board has a slower flame spread rate than the particle board based material, reflecting the ignition results. 126 Comparison of flame spread for melteca faced board 0 60 120 180 240 300 360 420 480 0 20 40 60 80 100 120 140 160 180 Distance along sample (mm) Ti m e f r om i g ni t i on ( s ) Melteca-MDF Melteca- particle board Figure 94: Flame spread of Melteca faced boards in RIFT The difference between the two board types is greater than that of the substrate alone, indicating that the surface material is having a greater effect, as seen in Figure 95, where the time to a point on the sample for Melteca faced boards (MDF and particle board substrate) and for the MDF and particle board (Pynefloor) alone. This could be a factor of the manufacturer of the boards, as the facing material was not controlled and the boards come from different manufacturers. 127 Comparison of flame spread on substrates(with and without Melteca facing) in RIFT - 60 120 180 240 300 360 420 480 - 60 120 180 240 300 360 420 480 MDF sheet or substrate (time to point) (s) P a r t i c l e - b o a r d sh eet o r su b s tr ate (s) . Melteca substrate MDF vs Particle board Figure 95: Comparison of flame spread time to point for MDF vs. particle board, with and without Melteca facing in RIFT The erratic nature of the burning behaviour gives inconsistent results for the flame spread measurements. The correlation shows a large scatter (Figure 96 and Figure 97), with the Melteca- particle board being noticeably worse. The greater sample size and resolution of the LIFT gives a better correlation than the RIFT for the same material. The data scatter in the RIFT results makes it more difficult to obtain accurate and consistent results for the flame spread correlation for Melteca faced boards. Tests with the Melteca faced boards without preheating gave no useable result ? there was no correlation of the data. Figure 98 shows the results for Melteca-MDF, where the data fit uses the measured value for the minimum ignition flux, and an estimated fit to the data and should be treated with caution. A mathematical fit to the data shows the slope of the best fit line being in the opposite direction. 128 Flame spread correlation for Melteca faced board in RIFT Melteca-PB y = -1.7756x + 110.11 R 2 = 0.0263 Meltec a- MDF y = -6.4377x + 197.88 R 2 = 0.4845 0 50 100 150 200 250 0 5 10 15 20 25 30 35 40 1/ ? V f (m / s ) -0 . 5 Melteca - particle board Metleca - MDF Figure 96: Flame spread correlation for Melteca faced board in RIFT Flamespread correlation for Melteca/MDF RIFT y = -6.4377x + 197.88 R 2 = 0.4845 LIFT y = -4.3019x + 141.96 R 2 = 0.8721 0 20 40 60 80 100 120 140 0 5 10 15 20 25 30 35 40 1/ ? V f ( m /s ) -0 . 5 RIFT LIFT Linear (RIFT) Linear (LIFT) Figure 97: Flame spread correlation for Melteca/ MDF 129 Flame spread correlation for Melteca/MDF in RIFT - low preheat 0 20 40 60 80 100 120 140 0 1 2 3 4 5 6 7 8 9 1011121314151617181920 1/ ? V f ( m /s ) -0 . 5 Data fit estimate based on measured q" ig,min value. Slope=-13.4 Figure 98: Flame spread correlation for Melteca-MDF with a low preheating period Flame spread of Melteca faced boards in LIFT The heat flux profile used in the LIFT test of Melteca- MDF is shown in Figure 99. The flame spread results for Melteca-MDF (Figure 97 and Figure 100) shows less data scatter than the same material in the RIFT. This is due to the larger scale of the LIFT test, making the variation due to the burning behaviour of the facing material less significant. 130 LIFT heat flux distribution for Melteca faced MDF 0 5 10 15 20 25 30 35 40 0 100 200 300 400 500 600 700 800 Distance along sample (mm) H e a t fl u x (k W / m 2 ) Figure 99: LIFT heat flux profile for Melteca faced MDF flame spread test Flame spread correlation for Melteca-MDF in LIFT Best fit y = -4.3019x + 141.96 R 2 = 0.8721 0 20 40 60 80 100 120 0 5 10 15 20 25 30 35 40 1/ ? V f ( m /s ) -0 . 5 Figure 100: Flame spread correlation for Melteca-MDF in LIFT 131 9 Material test results for natural timbers Natural timbers showed greater variation in the results between tests of identical materials than for manufactured boards, due to the greater variation in the material from the grain, knots or other features in the timber. The flame spread rates along the sample are generally higher for natural timbers than for the manufactured boards. 9.1 New Zealand Beech 9.1.1 Ignition of Beech The time to ignition for New Zealand Beech is shown in Figure 101.which shows greater variation in the time to ignition, than the other timber products (e.g. particle board Figure 54 and Figure 55) especially at low flux levels, Time to ignition vs heat flux for NZ Beech 0 120 240 360 480 600 720 840 960 1080 1200 1320 1440 1560 1680 10203040506070 Ti m e t o i g n i t i on t ig (s) LIFT ISO5657 RIFT Time limit for test in ASTMe1321-97a LIFT test Figure 101: Time to ignition vs. heat flux for NZ Beech The ignition parameter from ASTM E 1321-97a is shown below in Figure 102. While the ignition parameter results for the ISO 5657 and LIFT tests were similar at 0.050 s 0.5 for the ISO 5657 results, and 0.053 s 0.5 for the LIFT, the RIFT showed a wide 132 variation, due to the higher minimum ignition flux and shorter times to ignition than produced by the other methods. Ignition parameter for NZ Beech ISO = 0.0531x R 2 = 0.7253 LIFT = 0.0503x R 2 = 0.919 RIFT = 0.0344x R 2 = -0.5432 0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 0 5 10 15 20 25 30 35 40 45 ?(tig) (s 0.5 ) LIFT ISO5657 RIFT Figure 102: Ignition parameter for Beech The ISO 5657 and LIFT gave similar results for the critical ignition flux (Figure 103) and there is greater variation for the critical flux calculated from the RIFT ignition test. 133 Critical ignition flux for NZ Beech ISO = 0.0044x - 0.0526 R 2 = 0.9925 LIFT = 0.0042x - 0.0384 R 2 = 0.9777 RIFT = 0.0057x - 0.0725 R 2 = 0.8641 0.00 0.05 0.10 0.15 0.20 0.25 0.30 0 10203040506070 1/ ? (t ig ) (s -0 . 5 ) LIFT ISO5657 RIFT Figure 103: Critical ignition flux for NZ Beech 9.1.2 Flame spread of Beech Flame spread of NZ Beech in the RIFT The flame spread showed vide variations between runs in the RIFT, whether or not it was preheated prior to ignition (Figure 104 and Figure 105) and the flame spread in the RIFT was less consistent than in the LIFT (Figure 108). The low preheat time was 55 seconds, against a full preheat time for the RIFT of 844 seconds, and 395 seconds for the LIFT, using the ignition data from the LIFT ignition tests for this material. 134 Flame spread for 16mm NZ Beech in RIFT - low (55s) preheat time 0 60 120 180 240 300 0 50 100 150 200 250 Distance along sample (mm) Ti m e f r om i gni t i on ( s ) Test 1 Test 2 Test 3 Test 4 Test 5 Test 6 Figure 104: Flame spread of Beech in RIFT with 55 second preheating period The extent of flame spread is less if the material is not preheated and hence the experimental value for the minimum heat flux for flame spread is higher than for a fully preheated sample. As can be seen from time to reach each point on the sample in Figure 104, the rate of flame spread was very rapid in comparison to the other materials tested, indicating that the flame spread was faster than the surface of the material could reach the minimum ignition temperature, hence the extent of flame spread is less when the material has not been preheated fully prior to ignition. 135 Flame spread for Beech in RIFT - full preheat time 0 60 120 180 240 300 360 0 50 100 150 200 250 300 Distance along sample (mm) Ti m e f r o m i g ni t i on ( s ) Test 1 Test 2 Test 3 Test 4 Figure 105: Flame spread for Beech in RIFT with full preheat The effect that the ignition data has on the flame correlation is shown in Figure 106, where the same flame velocity measurements are used, but with the ?b? value and thermal equilibrium time ?t * ? from the ignition tests in the RIFT and ISO 5657 ignition apparatus. Flame spread correlation for NZ Beech - low (55s) preheating time RIFT ign y = -7.5996x + 80.027 R 2 = 0.7174 ISO ign y = -4.9302x + 80.027 R 2 = 0.7174 0 10 20 30 40 50 60 70 80 024681012141618 1/ ? V f (s -0. 5 ) ISO ign data RIFT ign data Figure 106: Flame spread correlation for NZ Beech in RIFT - low preheat time 136 The flame spread correlation for the RIFT with the full preheating period is compared with the LIFT in Figure 110. Flame spread of NZ Beech in the LIFT The heat flux profile used for the LIFT tests of NZ Beech is shown in Figure 107. Heat flux profile for LIFT tests for NZ Beech 0 5 10 15 20 25 30 0 100 200 300 400 500 600 700 800 Distance along sample (mm) Hea t f l u x ( k W / m 2 ) Figure 107: Heat flux profile for LIFT tests for NZ Beech The flame spread along the sample (Figure 108) showed much less variation than for the same material in the RIFT (Figure 105). 137 Flame spread for NZ Beech in LIFT 0 60 120 180 240 300 360 420 480 540 0 100 200 300 400 500 600 Distance along sample (mm) T i m e f r om i gnt i o n ( s ) Test 1 Test 2 Test 3 Figure 108: Flame spread for NZ Beech in LIFT The resulting flame spread correlation, shown in Figure 109 shows some curvature in the fit of the data, indicating that the material may not have reached thermal equilibrium, despite being preheated to the time given by the ignition results. Flamespread correlation for beech in LIFT y = -2.6644x + 61.304 R 2 = 0.8484 0 10 20 30 40 50 60 70 80 0 5 10 15 20 25 1/ sq r t (V ) Figure 109: Flame spread correlation for NZ Beech in LIFT The RIFT and LIFT gave similar results for the flame spread correlation (Figure 110) at 74.8 for the RIFT and 58.3 for the LIFT, although the RIFT has more scatter in the 138 data, shown by the lower R 2 value. The results for the correlated values for the minimum heat flux for flame spread and the minimum ignition flux are similar with both methods. Flame spread correlation for NZ Beech in RIFT and LIFT RIFT= -2.459x + 54.938 R 2 = 0.714 LIFT = -2.6644x + 61.304 R 2 = 0.8484 0 10 20 30 40 50 60 70 80 0 5 10 15 20 25 30 1/ ? V f ( s -0 . 5 ) RIFT LIFT RIFT LIFT q" s Figure 110: Flame spread correlation of Beech in RIFT and LIFT. 139 9.2 Macrocarpa 9.2.1 Ignition of Macrocarpa The three ignition test methods produced similar results for Macrocarpa (Figure 111), with the LIFT having longer times to ignition at low heat flux levels. Time to ignition for Macrocarpa 0 240 480 720 960 1200 1440 1680 1920 10203040506070 T i m e t o i gni t i on ( s ) LIFT ISO5657 RIFT Limit for ignition test of 1200s in ASTMe1321-97a Figure 111: Time to ignition for Macrocarpa The ignition parameter (Figure 112) and the critical ignition flux (Figure 113) show much more scatter in the data than the manufactured boards, such as MDF (Figure 46 and Figure 45) or particle board (Figure 56 - Figure 59). 140 Ignition parameter for Macrocarpa ISO = 0.0479x R 2 = -0.846 LIFT = 0.0589x R 2 = 0.7755 RIFT = 0.0555x R 2 = -0.1189 0.00 0.20 0.40 0.60 0.80 1.00 1.20 0 5 10 15 20 25 30 35 ?(t ig ) s 0.5 LIFT ISO5657 RIFT Figure 112: Ignition parameter for Macrocarpa Critical ignition flux for Macrocarpa LIFT = 0.0053x - 0.078 R 2 = 0.9609 ISO y = 0.0064x - 0.0963 R 2 = 0.929 RIFT y = 0.0064x - 0.099 R 2 = 0.9271 0 0.05 0.1 0.15 0.2 0.25 0.3 10203040506070 1/ ? (t ig ) (s -0. 5 ) LIFT ISO5657 RIFT Figure 113: Critical heat flux for Macrocarpa 141 9.2.2 Flame spread of Macrocarpa The resinous streaks in Macrocarpa can make the flame spread erratic, as the flame front can tend to follow a resin streak on the face of the timber, rather than spreading evenly over the face of the material. This gives more erratic flame spread than for a more consistent material, seen in the variation of the flame spread results between runs in Figure 115 and Figure 117. The preheating time was 40 seconds for the low preheating tests, and 436 seconds for a full preheating time in the RIFT, based on the ISO 5657 ignition tests, and 288 seconds for the LIFT ignition tests Flame spread in RIFT The flame spread for Macrocarpa showed wide variation between tests of the same material under the same conditions (Figure 114 and Figure 115), particularly towards the end of the sample. Flamespread for 16mm Macrocarpa in RIFT - low (40s) preheat time 0 60 120 180 240 300 360 420 480 0 50 100 150 200 250 Distance along sample (mm) Ti m e f r om i g n i t i o n ( s ) Test 1 Test 2 Test 3 Test 4 Test 5 Test 6 Figure 114: Flame spread of Macrocarpa in RIFT with low (40s) preheat time 142 Flame spread for 16mm Macrocarpa in RIFT with full preheat 0 60 120 180 240 300 0 50 100 150 200 250 Distance along sample (mm) T i m e f r om i gni t i on ( s ) Test 1 Test 2 Test 3 Test 4 Figure 115: Flame spread of Macrocarpa in RIFT ? full preheat of 436 seconds The resulting flame spread correlation for the RIFT results shows a lot of data scatter and the linear fit is a relatively poor match (Figure 116). The fit of the data, shown by the R 2 values seen on the chart, improves with preheating the sample prior to ignition, but is still poor. Flame spread correlation for Macrocarpa in RIFT Low preheat y = -7.7358x + 98.182 R 2 = 0.4179 Full preheat y = -2.9391x + 66.422 R 2 = 0.6474 0 10 20 30 40 50 60 70 80 90 100 110 120 0 5 10 15 20 25 1/ ? V f ( m /s ) -0 . 5 Low preheat Full preheat Low preheat Full preheat Figure 116: Flame spread correlation for Macrocarpa in RIFT 143 Flame spread of Macrocarpa in LIFT The flame spread for Macrocarpa in the LIFT showed similar variation between tests to the samples in the RIFT. Flame spread for 16mm Macrocarpa in LIFT 0 60 120 180 240 300 360 420 480 540 0 100 200 300 400 500 600 Distance along sample (mm) Tim e f r o m igni t i on ( s ) Test 1 Test 2 Test 3 Figure 117: Flame spread for Macrocarpa in LIFT The curve to the LIFT results in the flame spread correlation (Figure 118) indicates that the material was not preheated sufficiently, based on the work by Quintiere (1981), despite the preheating time coming from the LIFT ignition results. 144 Flame spread correlation for Macrocarpa in LIFT Best fit y = -2.5085x + 47.508 R 2 = 0.6125 0 10 20 30 40 50 60 70 80 0 5 10 15 20 25 1/ ? V f ( m /s ) -0 . 5 Figure 118: Flame spread correlation for Macrocarpa in LIFT The comparison of the RIFT and LIFT results (Figure 119) for the flame spread correlation shows the RIFT gives a higher value for the correlated value of the minimum ignition flux " min, . ig q although the flame spread modulus C (the slope of the best fit line) is similar. In this case, the value of the minimum flux for spread " . s q is higher for the RIFT than the LIFT. 145 Flame spread correlation for Macrocarpa RIFT= -2.9328x + 66.296 R 2 = 0.6755 LIFT = -2.5085x + 47.508 R 2 = 0.6125 0 10 20 30 40 50 60 70 80 0 5 10 15 20 25 1/ ? V f ( m /s ) -0 . 5 RIFT LIFT RIFT LIFT Figure 119: Flame spread correlation for Macrocarpa 146 9.3 Radiata Pine 9.3.1 Ignition of Radiata Pine The time to ignition for the three ignition test methods is given in Figure 120, where the ISO 5657 and RIFT gave similar results. As can be seen in the results, in some cases, the material still ignited after the limiting ignition test period. .During the course of the test, the material noticeably charred over the long test intervals, with some surface cracking. Localised surface hot spots formed, where an area around 2cm across would be glowing, when the rest of the sample showed no signs of heating. Heat flux vs time to ignition for Radiata Pine 0 120 240 360 480 600 720 840 960 1080 1200 1320 1440 1560 1680 10203040506070 Ti m e t o i gni t i on t ig (s) LIFT ISO5657 RIFT Time limit for test in ASTMe1321-97a LIFT test Figure 120: Time to ignition of Radiata Pine The three methods gave similar results for the critical ignition flux (Figure 121). 147 Critical ignition flux for Radiata Pine ISO = 0.0054x - 0.0643 R 2 = 0.9933 LIFT = 0.0044x - 0.0431 R 2 = 0.9886 RIFT = 0.0054x - 0.0592 R 2 = 0.9401 0 0.05 0.1 0.15 0.2 0.25 0.3 10203040506070 Incident heat flux q" e (kW/m 2 ) 1/ ? (t ig ) (s -0 . 5 ) LIFT ISO5657 RIFT Figure 121: Critical ignition flux for Radiata Pine The ignition parameter chart (Figure 122) shows that the three different methods gave widely different values for the equilibrium time t * despite eliminating the values for 8.0 " . " min, . < e ig q q for calculating the data fit. Ignition parameter for Radiata Pine LIFT = 0.0397x R 2 = 0.94 ISO = 0.0394x R 2 = 0 .5113 RIFT = 0.0434x R 2 = 0.3235 0 0.2 0.4 0.6 0.8 1 1.2 1.4 0 5 10 15 20 25 30 35 40 45 ?(t ig ) s 0.5 LIFT ISO5657 RIFT v Figure 122: Ignition parameter for Radiata Pine 148 9.3.2 Flame spread of Radiata Pine The low preheat time for the flame spread tests was 40 seconds. The full preheat time was 643 seconds in the RIFT and 341 seconds in the LIFT. Radiata Pine noticeably charred during the preheating period, and the area up to 180mm along the sample was difficult to ignite just by using the pilot flame. Flame spread of Radiata Pine in RIFT The flame spread along the Radiata Pine samples in the RIFT is shown in Figure 123 and Figure 124. The effect of preheating of the sample can be seen in Figure 125 - the preheated sample has a larger flame spread distance, with a lower time to reach each point on the sample, giving a higher flame front velocity. Flame spread for 16mm Radiata Pine in RIFT - low (40s) preheat 0 60 120 180 240 300 360 420 480 540 0 50 100 150 200 250 Distance along sample (mm) T i m e f r om i gni t i o n ( s ) Test 1 Test 2 Test 3 Test 4 Test 5 Test 6 Figure 123: Flame spread for Radiata Pine in RIFT ? low (40s) preheat 149 Flame spread for Radiata Pine in RIFT - full preheat 0 60 120 180 240 300 360 420 0 50 100 150 200 250 300 Distance along sample (mm) Tim e f r om ignit i o n ( s ) Test 1 Test 2 Test 3 Test 4 Figure 124: Flame spread for Radiata Pine in RIFT - full preheat The effect of preheating on the flame spread can be seen in Figure 125, where preheating increases the flame spread rate. Comparison of flame spread of Radiata Pine in RIFT 0 60 120 180 240 300 360 420 480 0 50 100 150 200 250 300 Distance along sample (mm) T i m e to p o i n t (s ) Full preheat Low preheat 2 std devn Figure 125: Comparison of flame spread of Radiata Pine 150 Flame spread correlation for Radiata Pine in RIFT - low preheat Best fit y = -12.271x + 125.4 R 2 = 0.4469 0 20 40 60 80 100 120 0123456789101121314 1/ ? V f ( m /s ) -0. 5 q" ig,min q" s Figure 126: Flame spread correlation for Radiata Pine in RIFT ? low preheat time Flame spread correlation for Radiata Pine in RIFT - full preheat Best fit y = -5.0101x + 88.025 R 2 = 0.6254 0 20 40 60 80 100 120 140 0 2 4 6 8 10 12 14 16 18 20 1/ ? V f (m / s )- 0. 5 Figure 127: Flame spread correlation for Radiata Pine in RIFT - full preheat Flame spread for Radiata Pine in LIFT The heat flux distribution for the flame spread tests in the LIFT is the same as that for Rimu, shown in Figure 135. 151 The flame spread rate (Figure 128) for Radiata pine shows less variation than some of the other material tested, such as Macrocarpa (Figure 117) or Beech (Figure 108). Flamespread of Radiata Pine in LIFT 0 60 120 180 240 300 360 420 0 100 200 300 400 500 600 Distance along sample (mm) T i m e f r om i g ni t i o n ( s ) Test 1 Test 2 Test 3 Figure 128: Flame spread of Radiata Pine in LIFT The more consistent flame spread behaviour (when compared to macrocarpa) gives a more accurate correlation for the flame spread (Figure 127) with less scatter in the data. Flame spread correlation for Radiata Pine in LIFT Best fit y = -3.9818x + 71.815 R 2 = 0.8034 0 10 20 30 40 50 60 70 0 2 4 6 8 101214161820 1/ ? V f ( m /s ) 0. 5 Figure 129: Flame spread correlation for Radiata Pine in LIFT 152 The RIFT gives a similar match for the flame spread correlation (Figure 130) but with more data scatter, shown by the lower R 2 value. Flame spread correlation for Radiata Pine in RIFT and LIFT RIFT = -5.0101x + 88.025 R 2 = 0.6254 LIFT = -3.9818x + 71.815 R 2 = 0.8034 0 20 40 60 80 100 120 140 0 5 10 15 20 1/ ? V f ( m /s ) -0 . 5 RIFT LIFT RIFT LIFT Figure 130: Flame spread correlation for RIFT and LIFT 153 9.4 Rimu 9.4.1 Ignition of Rimu The time to ignition for Rimu is shown in Figure 131. Rimu charred less than the softer woods such as Radiata Pine, but the samples tended to bow during the low heat flux tests, where the sample was exposed to long heating periods. It did not appear to have had a significant effect on the ignition results. Time to ignition for Rimu 0 120 240 360 480 600 720 840 960 1080 1200 1320 1440 1560 1680 10203040506070 T I m e t o ig n i t i o n t ig (s) LIFT ISO5657 RIFT q" ig,min Figure 131: Time to ignition of Rimu The resulting critical ignition flux (Figure 132) shows all the test methods gave similar results. The spread of the data was typical of the timber products tested. 154 Critical ignition flux for Rimu LIFT = 0.0042x - 0.0399 R 2 = 0.9844 ISO= 0.0045x - 0.0575 R 2 = 0.9915 RIFT = 0.005x - 0.0691 R 2 = 0.9524 0 0.05 0.1 0.15 0.2 0.25 10203040506070 1/ ? (t ig ) (s -0 . 5 ) LIFT ISO5657 RIFT Figure 132: Critical ignition flux for Rimu The fit of the ignition parameter for Rimu (Figure 133) required some data reduction. Most noticeably, the results for the RIFT ignition tests have significantly worse fit, given by the lower R 2 value, than those of the other methods. Ignition Parameter for NZ Rimu LIFT = 0.0492x R 2 = 0.9084 ISO = 0.0502x R 2 = 0.6169 RIFT = 0.0465x R 2 = 0.4961 0.00 0.20 0.40 0.60 0.80 1.00 1.20 1.40 0 5 10 15 20 25 30 35 40 45 ?(t ig ) s 0.5 LIFT ISO5657 RIFT LIFT ISO RIFT Figure 133: Ignition parameter of Rimu 155 9.4.2 Flame spread of Rimu Flame spread for Rimu in the RIFT There was insufficient flame spread in the RIFT with a full preheat period to get accurate data. The extent of flame spread was in the order of 130mm; however the initial 90-100mm was unusable, due to excessive charring. In this respect the extent of flame spread is similar to the case with a low preheat period (Figure 134). The low preheat period ranged from 67-80 seconds. The range was due to the time taken to get the sample to ignite after the preheating period. The full preheating period was 397 seconds, based on the ISO 5657 ignition data, or 395 seconds based on the LIFT ignition tests. Flame spread for 16mm NZ Rimu in RIFT with low preheat time 0 30 60 90 120 150 0 20 40 60 80 100 120 140 160 180 200 Distance along sample (mm) Ti m e f r om i gni t i on ( s ) Test 1 Test 2 Test 3 Test 4 Test 5 Figure 134: Flame spread along NZ Rimu in RIFT with a low preheat time of 67-80 seconds Flame spread for Rimu in the LIFT With the flux profile used in the flame spread tests in the LIFT, given in Figure 135, the resulting flame spread is given in Figure 136. The extent of flame spread in the LIFT results (Figure 136) shows significant variation, covering a range of over 125mm. Comparing this with the flux profile in Figure 135 gives a flux required for 156 flame spread ( " . s q ) of between 5-10kW/m 2 . This material would have benefited from more tests to give a more accurate estimate of the average value. The flux required for spread from the correlations (Figure 137 and Figure 138) is similar although generally lower than that based on the extent of flame spread. Both the RIFT and LIFT gave similar values for the correlated values for the minimum heat flux for flame spread, despite the lack of preheating in the tests with the RIFT. Flux profile for flame spread tests of Rimu and Radiata Pine in LIFT 0 5 10 15 20 25 30 0 100 200 300 400 500 600 700 800 Distance along sample (mm) H eat fl u x (kW / m 2 ) Figure 135: Flux profile of LIFT flame spread test for Rimu and Radiata Pine 157 Flame spread for 16mm NZ Rimu in LIFT 0 60 120 180 240 300 0 100 200 300 400 500 600 Distance along sample (mm) Ti m e f r om i g ni t i on ( s ) Test 1 Test 2 Test 3 Figure 136: Flame spread for Rimu in LIFT Flamespread correlation for Rimu in RIFT - low preheat time Best fit y = -4.7403x + 75.472 R 2 = 0.6079 0 10 20 30 40 50 60 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 1/ ? V f ( m /s ) -0 . 5 Figure 137: Flame spread correlation for Rimu in RIFT with ISO 5657 ignition ?b? value ? 67-90 seconds preheat time 158 Flamespread correlation for NZ Rimu in LIFT Best fit y = -2.1737x + 56.558 R 2 = 0.7798 0 10 20 30 40 50 60 0 5 10 15 20 25 30 35 1/ ? V f ( m /s ) -0. 5 Figure 138: Flame spread correlation for Rimu in LIFT 159 10 Discussion of results 10.1 Equipment specific issues 10.1.1 LIFT The tests conducted for this report were the first results from the new LIFT testing machine at the University of Canterbury. In general, after initial teething problems, it performed well; however there are some improvements which can be made for future work The limitation for the upper heat flux is dictated by the air supply. The air compressor which feeds the LIFT burner is also used for the air supply for other parts of the building. Using an 8cfm compressor in parallel with the main supply improved the consistency, and showed that over " 50 . mm q = 60kW/m 2 was possible; however the air supply was not consistent enough for extended running over 50kW/m 2 . There are a number of alternatives. Another compressor could be mounted in parallel to provided additional capacity, or the burner can be redesigned to run on a lower pressure. The current burner design, seen in Figure 13, uses a 25mm square supply backbone with 6mm feeder tubes to the rear of the burner box, and semi circular baffles and Kaowool inside the burner box to distribute the gas evenly across the burner face. Replacing this with a larger tube would allow the air supply pressure to be reduced and a forward curve centrifugal fan could then be used for the air supply. If this is controlled with a variable frequency drive with RPM readout, then the air supply volume can be calibrated to the fan speed. While the gas supply is controlled via a mass flow controller, the air supply is currently set with a variable pressure regulator and gauge, with a ball valve allowing additional restriction to be added to the regulator to stop fluctuation. The gauge is sufficient for crude adjustment, but the final mixture is set by the operator, based on the appearance of the flames from the panel and the required flux output. A highly precise adjustment has not proven to be necessary in practice, however easily repeatable settings would make the setup and changes to the flux levels faster and more consistent from run to run. This can be achieved with the current setup by using a 350-500 litre/min mass flow controller on the air supply, or changing to a variable speed fan with tachometer as described above. 160 Changing the burner system or modifying the manifold would also allow the burner to run at slightly lower flux levels than can currently be achieved. With flux levels less than " 50 . mm q = 25kW/m 2 the heating of the face of the burner becomes uneven, with a cooler section noticeable in the centre. An extra series of gas feed pipes into this area from the gas manifold backbone would help to reduce this problem. The measurement of the heat flux is critical to accurate results. Due to the narrower angle of 15 degrees when compared with the RIFT, as well as the larger scale, with a shallower flux profile, this tends to be less critical than the RIFT. The positioning of the holes in the LIFT measuring template are specified with a 2mm tolerance. This is easily achievable when making the template, however it was noticed with use that the holes tended to wear and chip. The flux gauge was bent to allow it to be clamped to the rail behind the sample holder. A dedicated holder which would ensure that the face of the flux gauge was parallel with the face of the measuring template and not canted would give more consistent results. As the 50mm position is used the most, making extra templates with just this hole or reinforcing this hole to prevent wear is recommended. While the calibration curve was within specification over the extent of the flame spread, it was outside the limit at the cold end of the sample. Some further adjustment of the apparatus may improve the match to the standard curve, however, this appears to be a common problem as it was reported by Nisted (1991) and Babrauskas and Wetterlund (1999). Pauner (2003) also reported the same issues with the IMO 653/ ASTM E 1317 round robin tests, which use the same equipment and calibration curve. At low flux levels, the errors in measurement become critical, and although the measurement results deviate from the standard value by a large percentage, the absolute value is low, and may approach the limits of accuracy of the measurements and the effect of variation from the burner. Robertson and Ohlemiller (1995) conducted tests to improve the consistency of measurements of the ASTM E 1317 marine finishes test at low flux levels. The error in measurements increased along the sample due to the effect of the convective boundary layer over the face of the sample. Their recommendation was that the face of the flux meter is proud of the surface of the measuring template to avoid the boundary layer. In practice, this is difficult as the refractory board is fragile and a thinner board will not last. The flux meter head 161 imposes a maximum thickness of the template board of 12mm in order for the face to be level with the template and not shielded inside the hole. The observation of the flame front in the LIFT is either directly via marks on the sample or via the observation mirror and viewing rake. Measuring the flame front by observing the progress directly is easier and more accurate, as the flame front is ill defined at the initial stages, and can be difficult to see in the mirror, and errors are compounded by the effect of parallax and the width of the pins. . A problem shared by both the RIFT and the LIFT for directly observing the flame front is the charring of the material during preheating. Lines marked in pen disappear, but those done with a graphite pencil can often still be distinguished even though the surface of the material has blackened. In order to increase the resolution and the number of data points collected, Dietenberger (1995) used a system where a pointer was moved via a hand crank parallel with the sample, in line with the flame front. The pointer was connected to a potentiometer and a data acquisition system, so that data collection is continuous and more automated with less chance of error. A modification for the University of Canterbury LIFT along these lines is recommended. 10.1.2 RIFT Some of the comments noted for the LIFT also apply to the RIFT. The issue of the template holes wearing is more pronounced on the RIFT, as the combination of the smaller scale, the rapid decay of the flux profile and steeper angle between the element face and the sample makes the setup and calibration both more difficult and more critical. The rapid decay of the heat flux along the sample makes location of the measuring holes critical, as a small change in the position can make a large change in the actual heat flux. It is reasonably straightforward to get the location of the sample holder and the angle consistent between different experiment setups by measuring from the face of the cone calorimeter element, and the changeover time from the cone calorimeter in the horizontal position to the RIFT is usually between 15-20 minutes. Modifying the apparatus to use templates and locating stops would reduce this source of error further and make the changeover faster. It is important when measuring the heat flux that the face of the heat flux gauge is parallel with the face of the template. This is easier than in the LIFT, as the operator can see the front face of the template. Due to the steeper 60? sample angle of the RIFT 162 against the 15? angle of the LIFT, small errors in holding the flux gauge in place can have a large effect. A mounting frame to provide consistency would reduce this error. The accuracy of the data can be increased by increasing the number of reading points, preferably to every 10mm, which would coincide with the data collection points along the sample. This is not possible with a single template, due to the proximity of the holes. It can be achieved using 3 templates, with measuring holes at 30mm centres, and each template is offset by 10mm from the other templates. Samples mounted in the LIFT are preset in a sample holder, which is slid into position. The RIFT as used in these experiments had a fixed frame, into which the sample and backing board was slid into place. This has the advantage of being easier to setup the sample, and makes tearing the foil backing less likely, and makes positioning the sample into place more consistent and faster. The RIFT sample holder frame is mounted on an angle bracket, which allows the frame to pivot to change the sample angle, without changing the distance from the cone element. This forms a flange along the bottom of the sample holder, seen in Figure 38. This affects the airflow over the face of the sample, seen visually during the ignition tests from the behaviour of the smoke emitted from the sample. The ledge disrupts the airflow, reducing the cooling from the convective airflow ? a flange is mounted on the bottom of the ASTM E 1621 ICAL intermediate scale calorimeter samples for the same reason (Babrauskas 2003). In the case of the RIFT, it decreases the time to ignition to close to that of the ISO 5657 ignition tests, where it would be expected to be higher due to the orientation. The minimum ignition flux is also increased. The effect of this is discussed in detail in the following sections; however it appears that the RIFT with the flange as currently constructed is not suitable for conducting the ignition tests which form part of the flame spread experiments. The charring of the initial part of the sample in the flame spread tests obliterates the surface marks on the face of the sample. A series of pins as markers (Figure 139) was used to monitor the flame spread rate. Generally it was found that if the surface was too heavily charred, so that the pencil marks were no longer visible, then flame spread would be too rapid for accurate measurement, or the material would no longer ignite. In addition the error due to parallax and the width of the pins was too large given the scale of the sample and measuring positions to be generally worthwhile 163 Figure 139: Marker pins in RIFT 10.2 Comparison of LIFT, ISO 5657 and RIFT ignition results The differences in the details between the apparatus, regarding sample size, orientation, pilot and heating source, etc are expected to affect the ignition results, and hence the values subsequently calculated. The RIFT has a ledge at the front as part of the mounting frame and this affects the convective flow across the face of the sample. The ASTM E 1623 ICAL intermediate scale calorimeter uses such a ledge to prevent convective cooling of the exposed sample surface, as the ignition time results are otherwise far greater than those produced by other methods (Babrauskas, 2003).As the radiatve component of the heat transfer coefficient is expected to be unaffected by the ledge, it appears that the RIFT as used in these experiments appears to have a lower convective coefficient h c than the LIFT. This gives a correspondingly reduced heat loss from the face of the sample, and consequently the RIFT has shorter ignition times than the LIFT. The ASTM E 1321-97a standard specifies that the LIFT has a heat transfer coefficient of 0.015kW/m 2 K, including the convective component, although Dietenberger (1996) found the heat transfer coefficient was much higher at the hot end of the sample. This 164 was due to the effect of turbulence induced by the gas burner, and was a function of the heat flux at the point, and the convection coefficient varied along the sample. The heat transfer coefficient for the cone calorimeter is approximately 0.010 kW/m 2 K (Janssens et al, 2003), and Babrauskas and Wetterlund (1999) used a value from other research of 0.0115kW/m 2 K. Babrauskas (2003) gives the convection coefficient for the cone calorimeter in the vertical position as 20% higher than in the horizontal position, or 0.012kW/m 2 K. Dietenberger (1995b) found that it was a function of the received heat flux, in the same manner as the LIFT. The RIFT would therefore be expected to have longer ignition times than the ISO 5657 apparatus, due to the orientation, in the same manner as the LIFT. Since the ignition times for the RIFT were similar to the ISO ignition test results, as discussed in more detail below in section 10.2.2, then this implies that the convection coefficient for the RIFT is similar to the ISO ignitability apparatus. The effect of the larger sample size in the LIFT is expected to be less of a cause for the difference in results. The main difference in the heat loss from the surface between a LIFT and RIFT sample is convective, and the convective coefficient h c ? (sample height) 1/4 (Babrauskas, 2003). 10.2.1 The effect of data reduction of ignition data As noted in section 3.4.2, the data reduction undertaken can greatly affect the results of the test, by affecting the preheating time t * and the ignition parameter b, from the slope of the fitted data points. As given in Equation (32), the thermal equilibrium time * t equals [1/(ignition parameter b )] 2 , a small change in the fit of the data points leads to excessively long or short preheating times. The effect of insufficient preheating is a curvature in the flame spread correlation, and a poorer linear data fit. The points which are discarded as being a poor fit are generally obvious, and the process given in section 3.4.2 appears to give consistent results. This is an area recognised by others (Babrauskas and Wetterlund, 1999) as an area which is unclear within the standard, and a definite procedure should be included in future revisions for consistency and clarity. 10.2.2 Time to ignition The time to ignition between the ISO 5657 apparatus and the LIFT (Figure 140) follows the expected behaviour shown by Dietenberger (1995), where the ignition 165 times are longer in the LIFT than the ISO 5657 test, although not as marked, as the average difference is 4.3% in these tests. While the difference is greater at lower flux levels, the scatter of the results is also greater, and this tends to hide the trend at flux levels less than 30kW/m 2 . This is also noted by Babrauskas (2003). Shields, Silcock and Murray (1993) investigated the time to ignition for particle board, plywood and spruce softwood in the horizontal and vertical positions in the ISO 5657 ignition apparatus and in the cone calorimeter, as well as inverted in the ISO 5657 apparatus. The cone calorimeter and the ISO 5657 apparatus produced similar ignition results. In all cases, where the sample was in the vertical orientation, the ignition times were significantly increased for flux levels between 20 and 40 kW/m 2 . Comparison of time to ignition between ISO5657 and LIFT ignition test 0 60 120 180 240 300 360 420 480 0 60 120 180 240 300 360 420 480 ISO ignition test (s) LI FT i g ni t i o n t e s t ( s ) Plywood MDF Melteca Particle board Hardboard Rimu Pine Macrocarpa Beech Average of all points Average overall = 4.3% longer igntition time in LIFT Figure 140: Comparison of time to ignition for ISO 5657 and LIFT ignition test 166 The trend is similar with the RIFT vs. the LIFT, with the LIFT having higher times to ignition than the RIFT (Figure 141). The time to ignition for the RIFT and ISO 5657 are similar (Figure 142) Comparison of RIFT and LIFT time to ignition 0 120 240 360 480 600 720 840 0 120 240 360 480 600 720 840 RIFT (s) LI F T ( s ) Plyw ood 17mm MDF Meltic a/MDF Chipboard - Pynefloor hardboard Rimu Pine Macrocarpa beech Figure 141: Comparison of time to ignition for tests in RIFT and LIFT 167 Comparison of RIFT and ISO5657 time to ignition 0 120 240 360 480 600 720 0 120 240 360 480 600 720 RIFT (s) I S O5657 (s ) Plyw ood - 17mm MDF Meltica/MDF Chipboard - Pynefloor hardboard Rimu Pine Mac r oc ar pa beech Figure 142: Comparison of time to ignition for tests in RIFT and ISO 5657 ignition apparatus 168 10.2.3 Thermal inertia k?c The thermal inertia is compared in Figure 143, where the differences between the types of apparatus are apparent. The ISO 5657 apparatus produces higher thermal inertial values than the LIFT. The RIFT shows much more variation and spread compared with the other methods, giving more inconsistent results when it is applied to calculate the flame spread parameter. Comparison of thermal inertia from ignition in LIFT, ISO5657 and RIFT - 0.50 1.00 1.50 2.00 2.50 P l y wo o d M DF Me lt ec a f ac ed MD F H a r d bo a r d NZ B e e ch R a dia t a p in e Ri m u M a c r o c arpa P y ne f lo or p ar ti cl e B o ar d Material T h e r m al i n er t i a k p c [( kW / m 2 K) 2 LIFT ISO5657 RIFT ign Figure 143: Comparison of thermal inertia (k?c) in LIFT, ISO 5657 and RIFT 10.2.4 Minimum ignition flux The minimum ignition flux is compared in Figure 144, where the ISO 5657 ignition test gives a consistently lower value for " min, . ig q than the LIFT. The RIFT is generally greater than or equal to the LIFT values. As a comparison, Dietenberger (1996) found that the minimum ignition flux for redwood was 4kW/m 2 greater in the LIFT when compared to the cone calorimeter in the horizontal position. Given that the ISO 5657 and cone calorimeter tests give similar results for the minimum ignition flux (Babrauskas, 2003), then the values from 169 ignition tests in the LIFT would be expected be higher than those obtained in the ISO 5657 ignition tests. This can be seen in Figure 144. The higher minimum ignition flux for the RIFT ignition, combined with the lower ignition times gives a higher ignition parameter value ?b?, which affects the final flame spread parameter ? value. Figure 144: Comparison of minimum ignition flux " min, . ig q 10.2.5 Comparison of correlation values and measured minimum ignition flux " min, . ig q The comparison of the correlated value and the measured value for the minimum ignition flux ( " min, . ig q ) is shown below for both the RIFT (Figure 145) and the LIFT (Figure 146). The values of the minimum ignition flux from the flame spread correlation, given in the procedure in Section 3.4.4, are consistently higher than the 170 experimental values. Babrauskas and Wetterlund (1999) also noted this finding with their results, where they used the LIFT for both the ignition tests and flame spread. Figure 145: Comparison of RIFT flame spread correlation and ISO 5657 ignition test results for minimum ignition flux " min, . ig q 171 Figure 146: Comparison of LIFT correlation and LIFT ignition test results for minimum ignition flux 172 10.2.6 Apparatus and flame spread It is apparent when looking at the flux distribution in the RIFT in Section 6.2, reproduced below in Figure 147, that the apparatus is limited to materials with a relatively low value for the minimum flux required for flame spread ( " . s q ), in order to get sufficient flame spread along the sample. Flux profile for RIFT - (60? sample angle, 850?C element, 45mm separation of sample to element 0 5 10 15 20 25 30 35 40 0 50 100 150 200 250 300 350 Distance along sample (mm) H eat f l ux ( k W / m 2 ) Figure 147: RIFT irradiance curve As the initial 80mm of most timber based samples in the RIFT was unusable for the flame spread test due to excessive charring during the preheat period, the practical limit in order to get sufficient data is around " . s q = 7kW/m 2 , giving flame spread measurements along to 180mm from the hot end of the sample. The rapid decay of the flux profile around this point makes the measurement of the flame spread velocity more difficult, as there is a significant change between each measuring point. Samples with a lower flame spread parameter ?, which consequently have a lower flame spread velocity, will give more accurate results. It is important that the markings on the sample match their position. A small variation of the actual position against the expected position (with the measured heat flux) will introduce errors in the RIFT measurements, due to the more rapid decay of the flux profile and the smaller resolution. As the LIFT has a shallower flux profile and a 173 longer sample area, it is not as critical in this regard. A tolerance of ?2mm is listed for the LIFT. The LIFT uses a sample holder which is slid into place against a stop, allowing it to maintain the positioning, whereas the RIFT as currently designed has the sample slide directly into a sample holder frame, giving less consistent positioning, due to the effect of the aluminium foil backing bunching at the ends of the sample. The separate frame allows this to be taken into account when the sample is setup in the sample, so it can be zeroed accurately prior to insertion. 10.2.7 Comparison of ignition parameter ?b? The ignition parameter ?b?, calculated from the ignition data, depends on the minimum ignition flux " min, . ig q , which has been shown to be dependent on the ignition apparatus, as well as the data reduction, as excluding data points can make a significant difference to the final value. As a result, the value of the ignition parameter can be expected to vary between the ignition methods used. Figure 148 shows that the data from the LIFT tests tends to give slightly higher ignition parameter values for the materials tested. This is expected to be largely due to the higher " min, . ig q values that this ignition test gives compared with the ISO 5657 test. 174 Comparison of ignition parameter "b" for ISO5657 and LIFT 0 0.01 0.02 0.03 0.04 0.05 0.06 0.07 0 0.01 0.02 0.03 0.04 0.05 0.06 0.07 ISO5657 (s -? ) LI F T ( s -? ) Plyw ood MDF Melteca faced MDF Hardboard NZ Beech Radiata pine Rimu Macrocarpa Pynefloor particle Board Figure 148: Comparison of ignition parameter "b" in ISO 5657 and LIFT 10.2.8 Flame spread parameter ? From equation (34), reproduced below in equation (46), the flame spread parameter depends on the ignition parameter ?b? value. As noted previously, this can vary depending on the data reduction used, and the experimental apparatus. The effect this has can be seen by a 10% change in the ignition parameter value in the range of the materials tested, which leads to approximately a 17% reduction in the resulting flame spread parameter. 2 )( 4 Cb ? ? = (46) 175 Generally the flame spread parameters calculated from the RIFT tests are higher than those from the same material in the LIFT (Figure 149). The results based on the ISO 5657 ignition data show less scatter than those on the RIFT ignition tests, reflecting the variation in the ignition results using the RIFT for ignition testing. Comparison of flame spread parameter between LIFT and RIFT (ISO ign data) 0 10 20 30 40 50 60 70 80 0 1020304050607080 LIFT (kW 2 /m 3 ) RI FT ( k W 2 /m 3 ) Plywood MDF Melteca faced MDF Hardboard NZ beech Radiata pine Macrocarpa Particle board Figure 149: Comparison between flame spread parameter for LIFT and RIFT (using ISO 5657 ign data for RIFT) 176 Comparison of flame spread parameter for LIFT and RIFT (using RIFT ignition data) 0 10 20 30 40 50 60 102030405060 LIFT (kW 2 /m 3 ) R I F T ( u s i ng R I FT i g n i ti on da ta ) (k W 2 /m 3 ) Plywood MDF Melteca faced MDF Hardboard Radiata pine Macrocarpa Particle board Figure 150: Comparison of flame spread parameter for LIFT and RIFT (using RIFT ignition data) 10.2.9 Minimum heat flux for flame spread A limiting factor to the RIFT test is the minimum flux required for flame spread ( " . s q ), as if this value is too high, then there is insufficient flame spread distance to give accurate results. The results are generally comparable for the LIFT and RIFT (Figure 151). The outliers of Rimu and Melteca faced MDF are largely because the flame spread stops in the area where the profile is rapidly diminishing with a change of 4kW/m 2 over the 25mm measuring interval of the flux measuring template, so a small error in the location can have a large effect on the final outcome. 177 Comparison of minimum flux for spread from extent of flame spread in LIFT and RIFT 0 2 4 6 8 10 12 14 16 0246810121416 LIFT (kW/m 2 ) RI F T ( k W / m 2 ) Plyw ood MDF Melteca faced MDF Hardboard NZ beech Radiata pine Rimu Macrocarpa Pynefloor particle board Figure 151: Comparison of minimum flux for flame spread " s q in LIFT and RIFT The results where there is a closer match are for materials where " . s q < 7kW/m 2 , giving a practical limit to the RIFT test for materials to get comparable results. This limit does depend on the flame spread behaviour of the material, and the effect of charring due to the preheating. If the flame spread rate is sufficiently low, and data can be obtained over the area that is normally too heavily charred to ignite easily, then it may be possible to get useable data above this level. The value for the experimental minimum heat flux for flame spread " . s q is still going to have significant error due to the rapid change in the flux profile in the RIFT in this area. Given that the minimum flux for spread is not known until the material is tested, an estimate of the likelihood of success of the RIFT flame spread test is necessary from the ignition tests. A comparison of the minimum ignition flux and the flux required for spread shows a poor correlation (Figure 152), however the products for which no flame spread results were possible (the Melteca faced boards, and Rimu) all have highest minimum ignition flux. This indicates there is an approximate limit for the 178 minimum ignition flux of " min, . ig q < 18 - 19kW/m 2 for a successful flame spread test in the RIFT. Minimum ignition flux from ISO5657 test vs. minimum flux for flame spread in RIFT 0 5 10 15 20 25 0 5 10 15 20 25 Minimum ignition flux q" ig,min (kW/m 2 ) M i n i m u m f l u x fo r sp r ead q " s (kW / m 2 ) Plywood MDF Melteca faced Particle board Melteca faced MDF Pynefloor Particle board flooring Superflake Particle board flooring Hardboard Beech Radiata pine Rimu Macrocarpa Figure 152: Minimum ignition flux from ISO 5657 test vs. minimum flux for spread for RIFT 10.3 Material and operational differences 10.3.1 Effect of preheating Quintiere (1981, 1983, 1984), and the resulting ASTM LIFT standard calls for the material to be preheated so that the surface reaches equilibrium, and hence the time transient function F(t) =1, simplifying the data reduction. It is also justified for fire modelling as the walls are expected to be heated by the fire in the compartment and it gives a more conservative approach as the calculated flame spread velocity will be higher. The process of preheating the material changes the material properties, such as drying the moisture from the wood, and charring or melting the material. These changes are not constant along the sample ? the charring of the sample occurs at the hot end of the sample, where as the cold end is close to room temperature. With excessive preheating times the material can char to such an extent that the material 179 can no longer be ignited. This problem occurred with hardboard in the RIFT, which had a longer preheating time than in the LIFT due to the results of the ISO ignition test. The material charred to such an extent that it cracked and fell out or self ignited, so the shorter LIFT preheating time was used. This issue also occurred for the tests conducted by Nisted (1991) where some materials would not ignite when tested with the standard preheating period, so only half the preheat period was used on those materials. Babrauskas (1999) reports that some of the materials tested, which covered a wider range of material types than these tests, gave more consistent results when no preheating period was used. The results from these tests indicate the opposite for these materials, with the flame spread rate and the correlations showing far more variation than when the material is preheated fully, for example, the results for particle board given in Figure 62 - Figure 65. The issue of charring of the material during the preheating period makes the first 80mm in the RIFT, and 150-200mm in the LIFT effectively useless for flame spread measurements. The material is above the minimum ignition flux, so that the flame spread travels across the material in this region almost instantly. The effect of preheating on the flame spread correlation and the slope of the data points, from which the flame spread parameter is calculated, is shown in Figure 153. The slope, which equates to the flame spread modulus C, decreases as the material is preheated until equilibrium is reached (i.e. F(t)=1), due to the scaling effect of the F(t) time transient function. 180 Effect of preheating on flame spread correlation for Pynefloor particle board in RIFT 0 20 40 60 80 100 120 140 160 0 5 10 15 20 25 1/ ? V f ( m /s ) -0. 5 Low preheat - 80s Half preheat - 385s Full preheat - 760s Figure 153: Effect of preheating on the flame spread correlation for Pynefloor in RIFT The preheat times for Figure 153 were approximately 80 seconds, 385 and 760 seconds, where 755 seconds was the calculated time for the surface to reach equilibrium from the ISO 5657 ignition tests. 10.3.2 Effect of substrates on Melteca and facings on boards It is apparent from the tests on the Melteca faced boards, outlined in Section 8.4.3 that having a layer with a higher minimum ignition flux over the comparatively more flammable substrate gives inconsistent results in the LIFT tests (, and these results are worse in the RIFT (Figure 97 and Figure 98) due to the smaller scale and more rapid decay of the flux profile. 10.3.3 Effect of thickness There is a limitation on the theory given in the ASTM E 1321 LIFT standard that the material is assumed to be thermally thick. As an illustration of the effect of the sample thickness on the ignition time, 9mm and 17mm plywood with a 12mm lightweight Kaoboard insulated backing board were compared with ignition in the RIFT (Figure 154). At high heat fluxes, the time to ignition is the approximately the same, regardless of the thickness, as the thermal penetration depth is less than the sample 181 thickness. As the time to ignition increases, the thermal penetration is greater than the thickness of the thin sample, and as the backing board is a good insulator, there is little heat loss from the back of the sample into the backing board, hence the sample tends to become thermally thin, the sample temperature increases and the ignition time is less. With the thicker sample, there is still heat loss into the unheated part of the sample, leading to a longer ignition time. Comparison of ignition times for plywood in RIFT ignition test 0 120 240 360 480 600 720 0 120 240 360 480 600 720 17mm plywood (s) 9 m m pl y w ood ( s ) Figure 154: Comparison of ignition times for plywood in RIFT test 182 10.4 Proposal for improved apparatus The objective of this research report was to investigate alternative methods to the ASTM LIFT for measuring flame spread properties. While the RIFT shows some success, the limitation on suitable materials in order to get sufficient and consistent flame spread data rules out the use of the RIFT for a lot of cases. During the fabrication and use of the LIFT, it became apparent that it would be possible to reduce the size and cost of the apparatus and increase the control of the radiant panel by using modern elements and thermocouple controllers. The flux profile over the first 150mm of the sample is almost constant, allowing the LIFT to be used ignition testing. If this section was eliminated, then the length of the sample can be reduced without reducing the resolution of the flame spread section. This can be achieved by changing the offset of the panel to the sample so that the level section of the flux profile is shorter. The panel size can therefore also be reduced. The ignition tests can still be done if the radiant panel can turn to be parallel with the face of the sample, and is on a slide, similar to the University of Canterbury LIFT, so that the spacing between the sample and the elements can be changed. The major change would be to use thermocouple controlled radiant electric elements. These are capable of reaching 1000?C within 3 minutes and give a black body radiant spectrum. The thermocouple controllers will give an output with far less variation than that of the gas panel, and the temperature across the face will be more uniform. One limitation is that the electrical elements cannot produce the same output as the gas panel, with the current supply of 20A/phase giving " 50 . mm q = 50kW/m 2 . This is primarily an issue for the ignition tests where higher heat fluxes are required. Having the element bank able to move closer to the sample will allow higher heat fluxes for ignition. This should allow for a testing method which is only slightly bigger than the ISO 5657 ignition testing apparatus. This compares with the 1.7m*0.9m footprint of the LIFT testing apparatus. The initial work is to make a full scale replacement of the LIFT gas panel, which can also be used as a reduced scale panel by eliminating some of the elements as required. 183 This allows the effect of scale to be determined, and would easily allow the use the current LIFT apparatus for direct comparison. The full scale radiant panel measures 490mm x 300mm and uses 15 Elstein SHTS elements. Each element measures 95mm square and is mounted at 100mm centres, and has a maximum continuous temperature of 900?C and an output of 800W. This would give a " 50 . mm q = 56kW/m 2 at the standard LIFT layout, as shown in Appendix 2. 184 11 Comparison of data with published literature There is little published data available from LIFT flame spread tests, reflecting the narrower focus of the test when compared to the cone calorimeter. As can be seen from the results, the variations between nominally similar materials can be significant if the preheating time is not adequate to get the material to equilibrium before ignition. Even within materials of the same type or source, the variation can be significant. There is a greater availability of ignition data for materials, some of which is summarised in Babrauskas (2003). Again, nominally identical materials can give different results for the same properties, depending on the testing apparatus, procedure and correlations or ignition theory used (Babrauskas, 2003) Comparisons have to be treated with caution, since apart from the materials used within the tests conducted for this report, they were not controlled. The materials often have a different base for the same nominal material. As an example the material basis is given as Douglas fir or Southern Pine for much of the literature which uses US construction plywood, and European plywood tests where the material details are specified are birch based. New Zealand plywood is generally Radiata Pine, other than specialty furniture and marine plywood. There is little data on NZ native timbers, other than that published by Ngu (2002) and Henderson (1998). The complete list of thermal properties and comparisons is given in Appendix 5. The list of properties from the literature is limited to those approximately the same as those tested for this report, in regard to material composition and surface finish. Not all the results can be compared, as much of the literature does not give the data for values such as " min, . ig q . 185 11.1 Comparison of Ignition results with published literature 11.1.1 Thermal inertia (k?c) The thermal inertia from the ignition tests generally compares well with the published data, as shown in Figure 155 - Figure 156 and Appendix 5 The thermal inertia plays a direct part in the flame spread calculation, given in equation (33). As can be seen from the results, there can be significant variation even with the same material, due to the correlations used and the effect of different testing methods. The detail of each material is given in Appendix 5 if it was specified in the original reference. The round-robin test report by Fowell of the LIFT apparatus, reported that a 2:1 spread of results was not uncommon, despite testing the same material, using the same methods in different laboratories. The thermal inertia for plywood could be expected to vary more than more homogeneous materials, due to the presence of voids in the material from bonding the ply. As this plywood was rated C/D grade, then some internal voids could be expected Comparison of effective thermal inertia for plywood 0 0.5 1 1.5 2 2.5 3 IS O 5 6 57 L I FT RIF T N g u, C K. 200 2 As a ka s an e t a l 199 8 Ji a n m i n, 1 990 Ni s t ed, 19 9 1 F ang r a t e t a l , 19 96 F an g ra t e t a l , 1 9 96 F an g ra t e t al , 1 9 96 F a ng ra t e t al , 19 9 6 F a ng r at e t al , 19 9 6 D ie t enb ur g e r, 1 9 9 5 (a ) Gr e xa , W h it e a nd Di e t en b er g er , 1 99 6 G r ex a , W h it e an d D ie te n be r ge r, 1 99 6 Gr e xa , W h it e a nd Di e t en b er ge r, 1 99 6 Qu int i e re & H a rk l e ro a d , 1 9 8 3 D ie te n bu rg e r, 19 9 5 ( c) F o w e ll , 1 993 ( N I ST ) Fo w el l , 19 93 ( UL ) Fo w e ll , 19 9 3 (S a fe ty E n g L a b) F o we ll , 1 9 93 ( N RC C) Source kp c (( kW / m 2 k 2 ) 2 Figure 155: Comparison of thermal inertia of plywood 186 Particle board and fibreboard products, such as MDF and hardboard show much more consistent results than plywood, due to the more homogeneous nature of the material. Despite being made from the same base material, there is a significant difference between the thermal inertia of Pynefloor and Superflake particle board and between the results from different test methods on the same material. Comparison of effective thermal inertia for particle board 0.00 0.50 1.00 1.50 2.00 2.50 Py n e fl o or IS O 5 65 7 Py n e f l o o r L I F T Py n e f lo o r RI F T S u pe r f l a k e I S O5 65 7 Su p e r f l a k e RI F T Di e t e n be rg er a n d G re x a , 19 9 9 Ja n ns en s M , 1 99 2 Qui n t ier e & H a r k l e r o ad , 19 83 B a br au s k as a n d W e t t e rl u n d , Cl ea ry & Q u in ti e re . 1 9 9 1 Di et e nb e rg er , 1 9 9 5 (a ) F a ng ra t e t a l , 1 9 9 6 Source kp c (( k W / m 2 k 2 ) 2 Figure 156: Comparison of thermal inertia of particle board 187 Comparison of effective thermal inertia for MDF - 0.20 0.40 0.60 0.80 1.00 1.20 1.40 1.60 ISO LIFT RIFT Ngu, CK. 2002 Asakasan etal 1998 Cleary & Quintiere, 1991 Henderson, A. 1998 Source kp c ((kW / m 2 k 2 ) 2 Figure 157: Comparison of thermal inertia of MDF Since the definition of ?hardboard? is not necessarily clearly defined, then some variation can be expected for this material (Figure 158). The results for Dietenberger come from a report on siding materials, so it may not be?hardboard? of the same type as tested in this report. 188 Comparison of effective thermal inertia of hardboard - 0.20 0.40 0.60 0.80 1.00 1.20 1.40 1.60 1.80 2.00 ISO LIFT RIFT Asakasan etal 1998 Dietenberger, 1995 (b) Dietenberger, 1995 (a) Quintiere & Harkleroad, 1985 from Quintiere, 2002 Quintiere & Harkleroad, 1985 from Quintiere, 2002 Source kp c ((k W / m 2 k 2 ) 2 Figure 158: Comparison of thermal inertia of hardboard When the thermal inertia for natural timbers is compared, there is much more variation. The value for thermal inertia from Moghtaderi et al. (1997) for Radiata Pine, shown in Figure 159, is for US grown specimens at 15% moisture content. The others are for New Zealand grown Radiata Pine, and the results for Henderson (1998) were at 11% moisture content, and both were measured in the cone calorimeter. Henderson (1998) got identical results for thermal inertia and ignition temperature for Rimu heart wood and sapwood, although the critical ignition flux was 20% higher for the heart wood over the sap wood. 189 Comparison of effective thermal inertia of radiata pine - 0.20 0.40 0.60 0.80 1.00 1.20 1.40 1.60 1.80 ISO LIFT RIFT Ngu, CK. 2002 Henderson, A. 1998 Moghtaderi, B et al. 1997 Source kp c ( ( k W /m 2 k 2 ) 2 (Cone calorimeter)(ISO5657) Figure 159: Comparison of thermal inertia of Radiata (Monterey) Pine Comparison of effective thermal inertia for macrocarpa and NZ rimu - 0.50 1.00 1.50 2.00 2.50 ISO LIFT RIFT Ngu, CK. 2002 Henderson, A. 1998 Source k ? c (k W / m 2 K) 2 macrocarpa Rimu Figure 160: Comparison of thermal inertial of Macrocarpa and NZ Rimu 190 11.2 Comparison of published flame spread results The significant effect of preheating on the results for the flame spread parameter has been covered previously in Section 10.3.1. The results from Huynh (2003) should be taken with care, as the correlation obtained by Huynh of the flame spread results was poor, and the material was not preheated. The results from Jinmen (1990) was calculated using the data given in the report by Jinmen as the flame spread parameter was not specifically stated as an outcome. 11.2.1 Flame spread parameter The results for plywood show that the flame spread parameter results in these tests on New Zealand Radiata based plywood are generally higher than the published values in the literature for plywood. The non-preheated results in the RIFT are comparable with those obtained by Huynh (2002), who also used Radiata pine plywood without preheating. Comparison of flame spread parameter ? for Plywood 0 20 40 60 80 100 120 140 160 RI F T F u ll p re h e a t - I SO i g n da t a R I FT Fu l l p r e h ea t - RI F T i g n d a t a RI F T Lo w p r eh ea t - I SO i g n da t a RI F T L o w p re h ea t - R I F T i g n d a ta L IFT Hy u n h , 20 0 2 P ea r c e (f r o m H y u n h , 2 0 0 2 ) As a k a s an e t a l 1 99 8 Ji a n m i n , 1 9 9 0 Ni s t e d , 1 99 1 Qu i n t ie r e & Ha rk le r o a d , 19 84 Fo we l l , 1 99 3 ( N I S T ) F o we l l, 19 9 3 (U L ) F o we l l , 1 99 3 ( S a f et y E n g L a b) Fo we l l , 1 9 9 3 ( NRCC ) Source ? (kW 2 /m 3 ) RIFT bas ed data LIFT based data Figure 161: Flame spread parameter for plywood 191 The results for medium density fibreboard (Figure 162) show that the flame spread parameter for this material is greatly affected by the preheating time. Fibreboard is an ill-defined material, as it can cover a range of materials made from wood fibre, including low density ceiling panel material as wall as medium and high density fibreboard, and the material specified as ?fibreboard? by Azhakesan et al (1998) is not clearly defined as MDF. No density information was available in the original publication, and a comment was made that the value they obtained was lower than expected, and this value was expected to be around 13 kW 2 /m 3 . Comparison of flame spread parameter for MDF 0 5 10 15 20 25 30 35 40 RIFT Full preheat - ISO ign data RIFT Full preheat - RIFT ign data RIFT Low preheat - ISO ign data RIFT Low preheat - RIFT ign data LIFT Asakasan et al, 1998 Hyunh, 2002 Cleary & Quintiere, 1991 Source ? (kW 2 /m 3 ) RIFT based data LIFT based data Figure 162: Comparison of flame spread parameter for MDF 192 Particle board is a commonly tested material in the LIFT literature, and much of the initial work on developing the theory was based on results for this material (Quintiere et al, 1981, 1983, 1984) The results for the RIFT are similar to the published data, but significantly less than the results from the LIFT with the same material. The results for a low preheating period are similar to Huynh (2002), which were also not preheated. The particle board used by Cleary (1991), Quintiere and Harkleroad (1984) and Babrauskas (1999) were Douglas Fir based, as opposed to the results from these tests, and Huynh (2002), which were all Radiata Pine based. Flame spread parameter ? for particle board 0 5 10 15 20 25 30 35 40 P / f l o o r R I FT Fu l l pr ehe at - I S O i g n P / f l o o r R I FT Fu l l pr ehea t - RI F T i g n P / f l oo r RI F T Low pr ehe at - I S O i g n P / f l oor RI F T Low pr ehea t - RI F T i g n P / f l oor LI F T S / f l a k e R I FT Fu l l pr ehe at - I S O i g n S / f l a k e R I FT Fu l l pr ehea t - RI F T i g n S / f l ak e RI F T Low pr ehe at - I S O i g n S / f l ak e RI F T Low pr ehea t - RI F T i g n Hy unh, 2 003 ( R I F T ) Q u i n t i er e & Har k l e r oad, 198 4 B abr a u s k a s and W e t t e r l und, 1999 Cl ear y & Q u i n t i er e . 1 991 Source ? (k W 2 /m 3 ) LIFT based data Figure 163: Flame spread parameter for particle board 193 12 Conclusions A full-scale ASTM E 1321 Lateral ignition and Flame Transport (LIFT) apparatus was constructed and a reduced scale ignition and flame spread apparatus (RIFT) was built to fit the cone calorimeter in the vertical position. A series of lateral flame spread tests with 9 different types of timber based products were conducted, and Quintiere?s flame spread model was applied to the results to obtain material properties. These materials included plywood, medium density fibreboard (MDF), hardboard, particle board flooring, Melteca covered MDF, Radiata Pine, NZ Rimu, Macrocarpa and New Zealand Beech. Further limited tests were conducted on Melteca covered particle board, and a second brand of particle board. As only timber based products were tested, the conclusions regarding the suitability of the RIFT can only be in relation to these materials. Other materials may give different results from those obtained in these tests. Test methodology These tests were based on the procedures given in the ASTM standard, and were conducted using common materials with both sets of equipment, so the results could be compared. Ignition tests were conducted in the LIT, RIFT and ISO 5657 ignitability apparatus. The flame spread tests in the RIFT were done with a preheating period, as required by the ASTM standard, and also without the preheating period, to gauge the effect of preheating on the results. The effect of sample preheating If the sample was not preheated to equilibrium before the flame spread test, the resulting values for the flame spread parameter are lower than with a fully preheated sample. This reflects the lower flame front velocity across the sample due to the greater difference between the surface temperature and the material ignition temperature. Non preheated samples also had more scatter in the correlation results, and the extent of flame spread was less than for preheated samples, if the flame spread was faster than the material in ahead of the flame front could reach equilibrium. This gives a higher value for the heat flux required for flame spread ( " . s q ) than is actually 194 achievable, hence in a fire situation, where some preheating can be expected, the flame front will spread further than anticipated. For comparable values to the LIFT, the testing method must be the same, so preheating is required in the RIFT in the same manner as the LIFT results. The reduced scale flame test method The RIFT method of flame spread testing has been shown to be successful for measuring flame spread properties; however, there are limitations on materials which can be reliably tested. These limitations are dictated by the smaller scale, such that the effects of errors on the results are more pronounced. The shape of the flux distribution profile along the sample has a sharp decay due to the directionality of the cone element, which leads to insufficient flame spread for accurate results for materials where the flux required for flame spread " . s q >7kW/m 2 . This equates to a maximum limit in the minimum flux for ignition " min, . ig q of 18-19 kW/m 2 , although this limit is not clearly defined. It was not possible to get flame spread results for Rimu, yet Macrocarpa gave acceptable results, despite both materials having the same value of " min, . ig q . Allied to the limitation of suitable materials dictated above, the material behaviour during flame spread is more critical. The smaller scale increases the effect of errors and erratic burning behaviour. The tests for Melteca covered MDF showed that the erratic behaviour caused buy the Melteca facing meant that no reliable correlation was possible in the RIFT, even though a result was obtained in the LIFT. Using the RIFT for the ignition tests proved to give inconsistent results and this most likely due to the design of the apparatus, which disrupts the airflow over the face of the sample. The ignition results which gave the best results when combined with the RIFT flame spread data came from the tests in the ISO 5657 ignitability apparatus. A view factor calculation was developed for the RIFT, based on earlier work by Huynh (2002), and compared against experimental findings. The simplifications in the view factor model led to results than were consistently to low against the experimental findings, although the match of the shape of the resulting curve was good. Flame spread parameter 195 The LIFT gives values for the ignition parameter ?b?, used in calculating the flame spread parameter values, which are generally higher than those obtained in the ISO 5657 ignition test, leading to higher values for the flame spread parameter ? for similar flame spread test results. The flame spread parameter ? calculated from the results in the RIFT, using the ISO ignition data is generally comparable to those obtained in the LIFT, although on average, the value tends to be higher than the value obtained in the LIFT. The more erratic ignition results for ignition tests in the RIFT give a wide variation in the flame spread parameter results, and this ignition method is not recommended. The values for the minimum flux for heat spread are comparable between the RIFT and the LIFT, for values which are under the recommended limit for successful RIFT tests. Ignition tests in the ISO 5657 apparatus generally gave faster times to ignition than the same material in the LIFT, with a difference of 4% overall. This difference is expected, but the magnitude of the difference is less than other tests in the literature. To get a more definitive result would require more ignition tests. Generally, the flame spread correlation gave the expected results for the minimum ignition flux against the experimental values. The correlated value is higher than the experimental values, in line with the findings of Quintiere et al, 1983. The University of Canterbury LIFT apparatus and future developments This report also gives the first results of the new LIFT testing machine, built at the University of Canterbury in the first half of 2005. After initial teething problems, it has proven to give a consistent output. The burner is running on LPG, rather than methane, originally specified in the ASTM standard, due to problems with supply of methane or natural gas. Due to the limitations imposed by LPG on the lowest flux levels possible by the burner, the burner position is adjustable using a 7/8? leadscrew to allow the burner to be positioned further away from the sample for lower flux levels during ignition tests. This has proven to be successful, achieving a minimum flux level at the sample face of 11kW/m 2 , and the lead screw allows for consistent positioning and fine adjustment. 196 The current air supply limits the burner output to a maximum sustained output of 50kW/m 2 at the sample face, although tests have shown that over 60kW/m 2 is possible if the air supply is available. A modification to the design is proposed, using a thermocouple controlled radiant electric panel instead of the gas burner, which will give a more consistent output, and greater control than the gas panel can provide. Allied to this is a proposed intermediate/ bench scale test apparatus, which will give the same sample resolution and flame spread testing flux profile, yet allow the use of a smaller and less expensive apparatus than the current LIFT. 197 References Anon. Elstein SHTS IR elements http://www.elstein.de/english/index_shts.htm.. Accessed 02-11-2005 Anon. Index of Species Information - species: Cupressus Macrocarpa. US Dept of Agriculture. US Forest Service. http://www.fs.fed.us/database/feis/plants/tree/cupmac/ Accessed 14-01-2006 Anon. Superflake materials safety data sheet. Fletcher Wood Panels, New Zealand. Issued 1 December 2004 Anon. Production of Planted Production Forest Rough Sawn Timber for Year Ended 31 March 2004 for All Species. Ministry of Agriculture and Forestry, New Zealand. http://www.maf.govt.nz/statistics/primaryindustries/forestry/production/annual-sawn- timber-highlights/aprodplantspec.htm. Accessed 14-11-2005 Anon; Pynefloor Materials Safety Data sheet. Fletcher Wood Panels, New Zealand. Issued 1 December 2004 http://www.fwp.co.nz/pdf/products/Pynefloor%20MSDS.pdf Azhakesan, A. Shields, T. Silcock, G. Ignition and opposed flow flame spread using a reduced scale attachment to the cone calorimeter. Fire Technology, Vol 14, No. 2. 1998. Azhakesan, A. Shields, T. Silcock, G. Combustibility parameters for enclosure lining materials obtained during surface flame spread using reduced scale ignition and flame spread technique. Fire Technology, Vol 34, No.3. 1998. Azhakesan, A. Shields, T. Silcock, G. On the nature, influence and magnitude of flame heat transfer during surface flame spread. Fire Safety Journal, Vol 35. pp189 - 222. 2000. ASTM E 1623 Standard Test Method for Determination of Fire and Thermal Parameters of Materials, Products, and Systems Using an Intermediate Scale 198 Calorimeter. American Society of Testing and Materials, ASTM International, West Conshohocken, PA. USA. 2004. ASTM E 1354. Standard test method for heat and visible smoke release rates for materials using an oxygen consumption calorimeter. American Society of Testing and Materials, ASTM International, West Conshohocken, PA. USA. 2004. ADJE1317. Adjunct to ASTM E 1321-97a. American Society of Testing and Materials, ASTM International, West Conshohocken, PA. USA. 2004. ASTM E 1317-97a. Standard Test Method for Flammability of Marine Surface Finishes. American Society of Testing and Materials, ASTM Iinternational, West Conshohocken, PA. USA. 1997 Babrauskas, V. Cone calorimeter annotated bibliography 1982-1991, technote 1296. National Institute of Standards and Technology. Gaithersburg, 1991. Babrauskas, V. Ten years of heat release research with the cone calorimeter. National Institute of Standards and Technology. Gathersburg, MD. USA. 1992. Also available from http://www.doctorfire.com/cone Babrauskas, V. Flame fluxes in opposed flow flame spread; a review of the literature. SP report 1995:06, SP Swedish National Testing and Research Institute, Sweden. 1995. Babrauskas, V., Ignition of Wood: A Review of the State of the Art. Interflam 2001, Interscience Communications Ltd., London. 2001. pp. 71-88 Babrauskas, V. The cone calorimeter. Section 3, Chapter 3. SFPE Handbook of Fire Protection Engineering, 3 rd edition. Editor Beyler C et al. Publ. NFPA, Quincy, MA. 2002. Babrauskas, V. Ignition Handbook. Fire Science Publishers, Fire Science and Technology, Inc, Issaquah, WA, USA. 2003 199 Babrauskas, V. Wetterlund, I. Comparative data from LIFT and cone calorimeter tests on 6 products, including flame flux measurements. SP report 1999:14, SP Swedish National Testing and Research Institute, Sweden. 1999 Beyler, Craig L. Hunt, Sean P. Lattimer, Brian Y. Prediction of ISO 9705 Room/Corner Test Results. Volume 1. CGR/DC-215-99-I. Coast Guard Research and Development Centre, Groton CT, USA. November 1999. http://stinet.dtic.mil/cgi- bin/GetTRDoc?AD=A378408&Location=U2&doc=GetTRDoc.pdf Cleary, T. Flammability Characterization with the LIFT Apparatus and the Cone Calorimeter. Fire Retardant Chemicals Association. Technical and Marketing Issues Impacting the Fire Safety of Building and Construction and Home Furnishings Applications. Spring Conference, 1992. Orlando, FL, Publ. Technomic Publishing Co., Lancaster, PA, 99-115 pp, 1992. http://fire.nist.gov/bfrlpubs/fire92/art046.html Cleary, T. Quintiere, J. Framework for Utilizing Fire Property Tests - NISTIR 4619 .International Association for Fire Safety Science. Fire Safety Science. Proceedings. 3rd International Symposium. July 8-12, 1991, Edinburgh, Scotland, Publ. Elsevier Applied Science, New York, Editors Cox, G.; Langford, B. pp 647-656, 1991. http://fire.nist.gov/bfrlpubs/fire91/art009.html. Comeford, J. The spectral distribution of radiant energy of a gas fired radiant panel and some diffusion flames. Combustion and Flame, vol 18, pp 125-132. 1972 Delichatsios, MA. A new interpretation of data from LIFT (Lateral Ignition and Flame Transport) apparatus and modifications for creeping flame spread. Interflam 99. Interscience Communications Ltd., London. 1999 Dietenberger, M. Protocol for Ignitability, Lateral Flame Spread, and Heat Release Rate Using Lift Apparatus. Nelson, G., editor. Fire and polymers II. Materials and tests for hazard prevention: Proceedings of 208th National meeting of the American Chemical Society; 1994 August 21-26; Washington, DC. ACS symposium series 599. 200 Washington, DC: American Chemical Society, 1995A. Chapter 29. http://www.fpl.fs.fed.us/documnts/pdf1995/diete95c.pdf Dietenberger, M. Experimental and analytical protocol for ignitability of common materials. Fire and Materials Journal, vol 19, pp89-94. 1995B. www.fpl.fs.fed.us/documnts/pdf1995/diete95b.pdf Dietenberger, M. Ignitability Analysis of Siding Materials Using Modified Protocol for Lift Apparatus. Fire and Materials, VOL. 20, 115-121. 1996A. www.fpl.fs.fed.us/documnts/pdf1996/diete96a.pdf Dietenberger, M. Ignitability analysis using the cone calorimeter and lift apparatus Proceedings of the International Conference on Fire Safety: July 22-26, 1996, Columbus, OH. Volume 22. pp 189-197. 1996B http://www.treesearch.fs.fed.us/pubs/8878 Drysdale, D. An introduction to fire dynamics, 2 nd edition. J Wiley & sons, Chichester, UK. March 2003. Fowell, A. Interlaboratory test program on ASTM E 1321: Standard test method for measuring material ignition and flame spread properties. Second edition, American Society of Testing and Materials, ASTM International, West Conshohocken, PA. USA. November 1994 Grand AF, Mehrafza, M. Evaluation of the effectiveness of fire resistant durable agents on residential siding using an ICAL based testing protocol. Fire and materials 2001. 7th international conference and exhibition. Proceedings. Interscience communications limited. January 22- 24, 2001, San Antonio, Texas, USA. pp 241-248, 2001. Goransson, U. Using the cone calorimeter to predict flame spread. Nordtest project 882-90. SP Report AR1991:32. Fire Technology, Boras, Sweden, 1990 201 Hagge, M. Bryden, K, Dietenberger, M. Effects of backing board materials on wood combustion. Wood & fire safety : proceedings, 5th international scientific conference, April 18-22, 2004 . Faculty of Wood Sciences and Technology, Technical University of Zvolen, Slovak Republic. pp51-58.2004. http://www.treesearch.fs.fed.us/pubs/7019 Henderson, A. Predicting Ignition Time Under Transient Heat Flux Using Results from Constant Flux Experiments. University of Canterbury, Christchurch, New Zealand . 1998 Hilado, C. Flammability test methods handbook. Technomic Publishing Co. Westport, Conn, USA. 1973 Howell, J. A catalog of radiation heat transfer view factors, 2 nd edition. http://www.me.utexas.edu/~howell/index.html. Accessed 14-01-2006 Huynh. CK. Flame spread measurements of New Zealand timber using a modified cone calorimeter. University of Canterbury, Christchurch, New Zealand. 2003. ISO 5657 ? 1986. Method of Measuring the Ignitability of Products Subjected to Thermal Irradiance. International Organisation of Standards, Geneva, Switzerland 1986 ISO5660. Reaction to fire tests ? heat release rate, smoke production and mass loss rate ? part 1. Cone calorimeter method. International Organisation of Standards, Geneva, Switzerland. 2002 ISO 9705. Full scale room fire test for surface finishes. International Organisation of Standards, Geneva, Switzerland. 1993 Janssens, M. Fire test and evaluation methods. APEC Timber and Fire conference, BRANZ New Zealand. Wellington 2005. http://www.branz.co.nz/branzltd/pdfs/MarcJanssens.pdf Janssens, M. L.; Kimble, J.; Murphy, D. Computer Tools to Determine Material Properties for Fire Growth Modelling From Cone Calorimeter Data. Fire and 202 Materials 2003. 8th International Conference. Conference Papers. Proceedings. .January 27-28, 2003, San Francisco, CA, pp 377-387, 2003. Interscience Communications Limited, New York. USA http://fire.nist.gov/bfrlpubs/fire03/art044.html Jianmin, Q. Prediction of flame spread test results using the test data from the cone calorimeter. SP Report 1990:38. Fire Technology, Boras, Sweden, 1990 Leung, C. Yuen W. Chow. W. A practical model of flame spreading over materials. Proceedings of the 6 th ASME-JSME Thermophysics Conference, March 16-23, Hawaii, USA. 2003. http://www.engineering.ucsb.edu/~yuen/current_paper/TED- AJ03-128.pdf Ngu. Ignition of New Zealand timber University of Canterbury, Christchurch, New Zealand. 2002 Nisted, T. Flame spread experiments in bench scale: project 5 of the Eurfic fire research programme. Dantest. Denmark. April 1991 Karlsson, B and Quintiere, J. Enclosure Fire Dynamics. CRC Press, Boca Raton, Florida. USA. 2000 Pauner, MA. Nordic Round Robin and Investigation on IMO Resolution A. 653 (16) TR 529. Nordtest. Finland. April 2003. Pease, T. ?A study of surface flame spread using a modified cone calorimeter?. Undergraduate final year project. Department of Chemical Engineering. University of Newcastle, Australia. 2001 Perrin, M. A comparison between cone calorimeter and LIFT flame spread data. Undergraduate final year project. Department of Chemical Engineering. University of Newcastle, Australia. 2002 203 Persson, B. On the prediction of lateral flame spread based on cone calorimeter data. SP Report 1993:28. Fire Technology, Boras, Sweden, 1993 Quintiere, J. A simplified theory for generalizing results from a radiant panel rate of flame spread apparatus. Fire and Materials, Vol 5, No. 2. 1981 Quintiere, J. Harkleroad. Walton, D. Measurement of flame spread properties. Combustion Science and Technology, Vol 32. pp 67-89. 1983 Quintiere, J. Harkleroad, M. New concepts for Measuring flame Spread Properties NBSIR 84-2943. National Institute of Standards and Technology. Gathersburg, MD. USA. November 1984. Quintiere, J. Principles of fire behaviour. Delmar publishers, USA. 1998 Quintiere, J. Surface flame spread. Chapter 12, Section 2, SFPE Handbook of Fire Protection Engineering, 3 rd edition. Editor Beyler C et al. Publ. NFPA, Quincy, MA. 2002 Rasbash, D. Ramachandran, G. Kandola, B. Watts, J. law. M. Evaluation of fire safety. J Wiley and Sons, Chichester, UK. 2004. Robertson, A. Ohlemiller, T. Low heat flux measurements: some predictions. Fire Safety Journal, Vol. 25, No. 2, 109-124, September 1995 Shields, T. Silcock,G. Murray, J. The effects of geometry and ignition mode on ignition times obtained using a cone calorimeter and ISO ignitability apparatus. Fire and Materials, Vol 17, pp 25-32. 1993 Shields TJ. Silcock, G. Murray, J. Evaluating ignition data using the flux time product. Fire and Materials, Vol 18. pp 243-354. 1993 Spearpoint, M. Huynh, M. Moghtaderi, B. Merryweather, G. Flame spread measurements of New Zealand timber using an adaptation of the cone calorimeter. 204 APEC Timber and Fire conference, Wellington, New Zealand, 2005. http://www.branz.co.nz/branzltd/pdfs/MichaelSpearpoint.pdf Smith, S. Effects of moisture on combustion characteristics of live California chaparral and Utah foliage. A thesis submitted to the faculty of Brigham Young University in partial fulfilment of the requirements for the degree of Master of Science. Department of Chemical Engineering Brigham Young University, Utah, USA. August 2005 Simpson, W and TenWolde, A. Physical properties and moisture relations of wood. Ch. 3. Wood handbook - Wood as an engineering material. Gen. Tech. Rep. FPL- GTR-113. Forest Products Laboratory Madison, WI: U.S. Department of Agriculture, Forest Service, Forest Products Laboratory. 1999 White, R and Dietenberger, M. Fire Safety. Ch 17, Wood handbook - Wood as an engineering material. Gen. Tech. Rep. FPL-GTR-113. Forest Products Laboratory Madison, WI: U.S. Department of Agriculture, Forest Service, Forest Products Laboratory. 1999 White, R and Dietenberger, M. Cone Calorimeter evaluation of wood products. 15th Annual conference on flame retardancy 2004. http://www.fpl.fs.fed.us/documnts/pdf2004/fpl_2004_white002.pdf (accessed Jan 2006) Wilson MT; Dlugogorski BZ and Kennedy EM, ?Uniformity of Radiant Heat Fluxes in Cone Calorimeter?, Process Safety and Environmental Protection Group, The University of Newcastle, Callaghan, Australia, 7th Symposium in Fires Science in Worcester, USA, 2002. Younquist, J. Wood based products and panel products. Chapter 10. Wood handbook - Wood as an engineering material. Gen. Tech. Rep. FPL-GTR-113. Forest Products Laboratory, Madison, WI: U.S. Department of Agriculture, Forest Service, Forest Products Laboratory. 1999 205 Appendix 1: Procedure for operating University of Canterbury LIFT The gas supply to the main radiant panel is controlled by a mass flow controller, and the compressed air supply is adjusted through an air regulator and ball valve (Figure 164). The ball valve allows the regulator to be used in the optimum range by providing a bulk adjustment to the output restriction. Without the ball valve set to give a minimum of 6-8 psi on the gauge, the regulator will fluctuate. The limiting flux output without preloading the regulator is where " 50 . mm q < approximately 26kW/m 2 . Closing the ball valve part way loads up the regulator to 6-8psi and gives a stable output. The pilot burner uses a mass flow controller for the air supply, and a separate LPG supply. Starting the LIFT panel ? Insert the measurement template into the sample holder and fit the flux gauge in the 50mm position, so the face of the flux gauge is level and parallel with the face of the template. ? Turn on the water supply to the flux gauge. ? Close the ball valve so the air is turned off. ? Turn on gas and ignite the panel and set the gas output to the desired value on the mass flow controller by adjusting the set point adjustment (Figure 165). ? Slowly open the ball valve and adjust the air supply with the regulator and ball valve so that the regulator reads a minimum of 8 psi. Leave the mixture slightly rich, shown by the yellow flame tips, until the temperature and flux output has stabilised. As the burner heats up, it tends towards a leaner gas mixture and can flash back inside the burner. Adjustments are made by either increasing the gas flow or decreasing the air so that gas mixture is rich during the flux alteration, so it doesn?t flash back inside the burner. Once the burner is stable, which usually takes around 5 minutes, then the gas mixture can be adjusted for a neutral mix. Monitor the flux meter for a further time of approximately 5 minutes to ensure that it is stable. Extinguish the radiant panel by turning off the air supply first and then the gas supply 206 Figure 164: Regulator and ball valve. LIFT settings for gas supply and air restriction ball valve 0% 20% 40% 60% 80% 100% 120% 0 102030405060 q" 50 Heat flux (kW/m 2 ) Gas a n d b a l l v a l ve set t i n g % 0 2 4 6 8 10 12 14 16 18 20 A i r p r essu r e (p si ) Gas controller Ball valve Air pressure Figure 165: LIFT settings Setting the pilot flame Set the air supply flow rate to minimum and turn off the pilot air supply mass flow controller and turn on the pilot burner gas supply and ignite the pilot flame. Turn on the air supply and adjust the gas needle valve and regulator and air supply to give a blue pilot flame approximately 180mm long. The flame mixture usually has to be slightly rich to be stable with the air turbulence during an ignition test. Once the pilot flame is set, further adjustment is not usually required. 207 Appendix 2: Electrical power and current requirements for an electric LIFT radiant panel Angle of sample to element 15 deg Flux at 50mm point on sample 50.5 kW/m 2 Standard LIFT panel dimensions: Burner 140 301.8 181.2 140 208 From: Howell, http://www.me.utexas.edu/~howell/index.html Area of panel segment (in line with 50mm point) A1 = A2 0.042252 m 2 a3=a4 0.025368 m 2 Panel width 0.28 m Panel length 0.483 m Element temp 1140.5 K Stefan-Boltzman const. ? 5.67E-08 Wm 2 /K 4 Emissivity ? 0.99 (from Wilson et al for a cone calorimeter element) " 2:1 . AA q =F. ??.(T) 4 a 0.140 m b 0.302 m c 0.172 m 209 F (A1,A2) (View factor) 0.147 " 2:1 . AA q 13979 W/m 2 " 4:3 . AA q ) =F.??(T) 4 a 0.14 m b 0.181 m c 0.172 m F (A3,A4) (View factor) 0.128 " 4:3 . AA q 12143 W/m 2 " . e q = 52243 Radiation received at point (angled 15 deg) 50463 W/m 2 received at point Radiation on sample 50.5 kW/m 2 at 50mm sampling point Element emitted heat flux 95.0 kW/m 2 emitted from surface of element Element area 0.135 m 2 Element power 12.8 kW current (240v) 54 Amps if electrically powered at 240v 210 Appendix 3: Results of material tests Ignition test results MDF ? 18mm LIFT ISO 5657-1986 Ngu (ISO 5657) Ignition in RIFT Ignition parameter b 0.047 0.053 0.043 0.053 s 0.5 intercept of ignition parameter graph (t*) 0.5 21 19 23 19 s 0.5 Preheat time t* 456 355 544 357 sec Critical ignition flux " . crit q 3.4 3.7 4.7 -0.3 kW/m 2 Thermal inertia kpc 1.10 0.86 1.04 0.86 (kW/m 2 K) 2 Minimum ignition flux " min, . ig q 16.25 16.25 13.5 16.25 kW/m 2 Heat transfer coefficient at ignition h 0.044 0.044 0.039 0.043 kW/m 2 K Ignition temperature T ig 391 391 348 391 ?C Plywood - 17mm LIFT ISO 5657-1986 Ngu (ISO 5657) Ignition in RIFT Ignition parameter b 0.058 0.048 0.034 0.061 s 0.5 intercept of ignition parameter graph (t*) 0.5 17 21 29 17 s 0.5 Preheat time t* 293 432 870 273 sec Critical ignition flux " crit q 7.2 4.6 7.2 9.9 kW/m 2 Thermal inertia kpc 0.71 0.93 1.76 0.66 (kW/m 2 K) 2 Minimum ignition flux " . crit q 16.3 13.8 12.0 16.3 kW/m 2 211 Heat transfer coefficient at ignition h 0.044 0.041 0.040 0.043 kW/m 2 K Ignition temperature T ig 391 352 321 391 ?C Melteca faced MDF LIFT ISO 5657-1986 Ignition in RIFT Ignition parameter b 0.0423 0.0429 0.051 s 0.5 intercept of ignition parameter graph (t*) 0.5 24 23 19.5 s 0.5 Preheat time t* 560 542 382 sec Critical ignition flux " . crit q -6.44 2.58 1.06 kW/m 2 Thermal inertia kpc 1.475 1.474 1.07 (kW/m 2 K) 2 Minimum ignition flux " min, . ig q 18.75 20 21.25 kW/m 2 Heat transfer coefficient at ignition h 0.045 0.046 0.047 kW/m 2 K Ignition temperature T ig 440 425 454 ?C Melteca/ Particle board LIFT ISO 5657-1986 Ignition in RIFT Ignition parameter b Not tested 0.044 0.057 s 0.5 intercept of ignition parameter graph (t*) 0.5 Not tested 22.6 17.5 s 0.5 Preheat time t* Not tested 512 308 sec Critical ignition flux " . crit q Not tested 2.00 10.1 kW/m 2 Thermal inertia kpc Not tested 1.27 0.97 (kW/m 2 K) 2 Minimum ignition flux " min, . ig q Not tested 18.75 23.75 kW/m 2 Heat transfer coefficient at ignition h Not tested 0.044 0.050 kW/m 2 K Ignition temperature T ig Not tested 425 478 ?C 212 Pynefloor particle Board LIFT ISO 5657-1986 Ignition parameter b 0.0509 0.0364 0.063 s 0.5 intercept of ignition parameter graph (t*) 0.5 19.6 27.5 16 s 0.5 Preheat time t* 385.6 754.6 250 sec Critical ignition flux " . crit q 1.7 3.5 4.3 kW/m 2 Thermal inertia kpc 0.96 1.64 0.70 (kW/m 2 K) 2 Minimum ignition flux " min, . ig q 18.75 13.75 21.25 kW/m 2 Heat transfer coefficient at ignition h 0.044 0.041 0.047 kW/m 2 K Ignition temperature T ig 425 352 454 ?C Superflake particle board LIFT ISO 5657-1986 Ignition in RIFT Ignition parameter b Not tested 0.0327 0.05 s 0.5 intercept of ignition parameter graph (t*) 0.5 Not tested 30.6 19.72 s 0.5 Preheat time t* Not tested 935 389 sec Critical ignition flux " . crit q Not tested 3.44 5.24 kW/m 2 Thermal inertia kpc Not tested 2.11 1.05 (kW/m 2 K) 2 Minimum ignition flux " min, . ig q Not tested 13.75 18.75 kW/m 2 Heat transfer coefficient at ignition h Not tested 0.042 0.05 kW/m 2 K Ignition temperature T ig Not tested 352 424.87 ?C 213 Hardboard LIFT ISO 5657-1986 Ignition in RIFT Ignition parameter b 0.052 0.033 0.032 s 0.5 intercept of ignition parameter graph (t*) 0.5 19 31 31.3 s 0.5 Preheat time t* 375 944 977 sec Critical ignition flux " . crit q 2.172 - 0.588 2.6 kW/m 2 Thermal inertia kpc 0.88 1.83 1.67 (kW/m 2 K) 2 Minimum ignition flux " min, . ig q 17.5 11.25 11.25 kW/m 2 Heat transfer coefficient at ignition h 0.043 0.039 0.037 kW/m 2 K Ignition temperature T ig 409 307 307 ?C NZ Beech LIFT ISO 5657-1986 Ngu (ISO 5657) Ignition in RIFT Ignition parameter b 0.050 0.053 0.025 0.034 s 0.5 intercept of ignition parameter graph (t*) 0.5 20 19 40 29.1 s 0.5 Preheat time t* 395 355 1,618 844 sec Critical ignition flux " . crit q 9.24 11.83 11.14 12.8 kW/m 2 Thermal inertia kpc 0.98 0.85 3.27 2.30 (kW/m 2 K) 2 Minimum ignition flux " min, . ig q 18.75 18 12 18.75 kW/m 2 Heat transfer coefficient at ignition h 0.044 0.043 0.040 0.046 kW/m 2 K Ignition temperature T ig 425 415 321 425 ?C 214 Radiata Pine LIFT ISO 5657-1986 Ngu (ISO 5657) Ignition in RIFT Ignition parameter b 0.054 0.039 0.047 0.043 s 0.5 intercept of ignition parameter graph (t*) 0.5 18 25 21 23 s 0.5 Preheat time t* 341 643 460 531 sec Critical ignition flux " . crit q 9.7 12.0 7.8 10.9 kW/m 2 Thermal inertia kpc 0.92 1.54 1.08 1.45 (kW/m 2 K) 2 Minimum ignition flux " min, . ig q 18.75 16.25 15.5 18.75 kW/m 2 Heat transfer coefficient at ignition h 0.046 0.043 0.043 0.046 kW/m 2 K Ignition temperature T ig 425 391 380 425 ?C Rimu LIFT ISO 5657-1986 Ngu (ISO 5657) Ignition in RIFT Ignition parameter b 0.050 0.050 0.033 0.047 s 0.5 intercept of ignition parameter graph (t*) 0.5 20 20 30 21.5 s 0.5 Preheat time t* 395 397 893 462 sec Critical ignition flux " . crit q 9.62 12.75 7.52 13.9 kW/m 2 Thermal inertia kpc 1.06 1.03 1.93 1.41 (kW/m 2 K) 2 Minimum ignition flux " min, . ig q 18.5 18 13.5 21.25 kW/m 2 Heat transfer coefficient at ignition h 0.046 0.045 0.041 0.049 kW/m 2 K Ignition temperature T ig 422 415 348 454 ?C 215 Macrocarpa LIFT ISO 5657-1986 Ngu (ISO 5657) Ignition in RIFT Ignition parameter b 0.059 0.048 0.055 0.056 s 0.5 intercept of ignition parameter graph (t*) 0.5 17 21 18 18.0 s 0.5 Preheat time t* 288 436 337 325 sec Critical ignition flux " . crit q 14.7 15.2 12.9 15.6 kW/m 2 Thermal inertia kpc 0.78 1.13 0.79 0.89 (kW/m 2 K) 2 Minimum ignition flux " min, . ig q 18.75 18 15.5 18.75 kW/m 2 Heat transfer coefficient at ignition h 0.046 0.045 0.043 0.046 kW/m 2 K Ignition temperature T ig 425 415 380 425 ?C 216 Ignition test data Plywood - 17mm ISO 5657 RIFT LIFT Heat flux (kW/m 2 ) Time to ignition (s) Heat flux (kW/m 2 ) Time to ignition (s) Heat flux (kW/m 2 ) Time to ignition (s) 10 no ignition 12.5 No ign 12.5 No ign 12.5 1690 15 No ign 15 No ign 15 1125 17.5 663 17.5 760 20 229 20 611 20 173 30 69 30 82 30 114.7 40 49 40 52.8 40 42.4 50 28.7 50 19.5 50 26 60 23.3 60 18.5 Medium Density fibreboard (MDF) ISO 5657 RIFT LIFT Heat flux (kW/m 2 ) Time to ignition (s) Heat flux (kW/m 2 ) Time to ignition (s) Heat flux (kW/m 2 ) Time to ignition (s) 15 no ignition 17.5 343 15 No ign 17.5 723 20 215 17.5 760 20 239 30 92 20 173 30 101 40 60.5 30 114.7 40 56.4 50 30 40 42.4 50 39.2 60 28 50 26 60 25.4 217 Melteca / MDF ISO 5657 RIFT LIFT Heat flux (kW/m 2 ) Time to ignition (s) Heat flux (kW/m 2 ) Time to ignition (s) Heat flux (kW/m 2 ) Time to ignition (s) 20 No ign 20 No ign 17.5 No ign 21 977 22.5 349 20 401 22.5 467 30 188 30 257 25 354 40 108 40 153 30 204 50 58.1 50 93.5 40 126 60 53.4 50 84 60 62 Melteca / particle board shelving ISO 5657 RIFT LIFT Heat flux (kW/m 2 ) Time to ignition (s) Heat flux (kW/m 2 ) Time to ignition (s) Heat flux (kW/m 2 ) Time to ignition (s) 17.5 No ign 22.5 No ign Not tested 20 558 25 292 22.5 356 30 221 25 255 40 95 30 170 50 54.2 35 128 60 30.2 40 93 50 59 60 45.4 218 Pynefloor particle board ISO 5657 RIFT LIFT Heat flux (kW/m 2 ) Time to ignition (s) Heat flux (kW/m 2 ) Time to ignition (s) Heat flux (kW/m2) Time to ignition (s) 12.5 No ign 20 No Ign 17.5 No ign 15 1166 22.5 226 20 315 17.5 575 30 140 30 175 20 368 40 64.8 40 86.5 25 215 50 35 50 50.5 30 154 60 27.5 35 98 40 93 50 47 60 38.4 Superflake particle board ISO 5657 RIFT LIFT Heat flux (kW/m 2 ) Time to ignition (s) Heat flux (kW/m 2 ) Time to ignition (s) Heat flux (kW/m 2 ) Time to ignition (s) 12.5 No ign 17.5 No ign Not tested 15 1090 20 404 17.5 525 30 123 20 333 40 75 25 210 50 33 30 128 60 39.4 35 88 40 75 50 48 60 31.4 219 Hardboard ISO 5657 RIFT LIFT Heat flux (kW/m 2 ) Time to ignition (s) Heat flux (kW/m 2 ) Time to ignition (s) Heat flux (kW/m 2 ) Time to ignition (s) 10 1378 12.5 886 15 1403 12.5 712 15 605 17.5 1223 15 402 17.5 338 20 259 20 285 20 273 30 108.5 30 127 30 117 40 96 40 74 40 62 50 64.5 50 52 50 30 60 35 60 28.4 Rimu ISO 5657 RIFT LIFT Heat flux (kW/m 2 ) Time to ignition (s) Heat flux (kW/m 2 ) Time to ignition (s) Heat flux (kW/m 2 ) Time to ignition (s) 17.5 No ign 20 No ign 15 1549 18.5 1358 22.5 544 17.5 1308 20 1134 30 219.5 20 366 30 186 40 53 30 187.5 40 61 50 24.1 40 61 50 31.5 60 22.4 50 34.4 60 24 220 Pine ISO 5657 RIFT LIFT Heat flux (kW/m 2 ) Time to ignition (s) Heat flux (kW/m 2 ) Time to ignition (s) Heat flux (kW/m 2 ) Time to ignition (s)) 15 No ign 17.5 No ign 15 1490 17.5 1154 20 660 17.5 1348 18.5 982 30 96.6 20 331 20 586 40 37 30 131 30 109 50 16.9 40 61 40 43 60 17.5 50 29.5 50 21.3 60 16.6 Macrocarpa ISO 5657 RIFT LIFT Heat flux (kW/m 2 ) Time to ignition (s) Heat flux (kW/m 2 ) Time to ignition (s) Heat flux (kW/m 2 ) Time to ignition (s)) 17.5 No ign 17.5 No ign 17.5 No ign 18.5 1767 20 976 20 1439 20 1077 30 182 30 114 30 237 40 43.6 40 74.5 40 32.5 50 14.3 50 26 50 14.2 60 15.4 60 15.9 221 NZ Beech ISO 5657 RIFT LIFT Heat flux (kW/m 2 ) Time to ignition (s) Heat flux (kW/m 2 ) Time to ignition (s) Heat flux (kW/m 2 ) Time to ignition (s) 17.5 glowing at 1800 20 No ign 17.5 No ign 18.5 1526 22.5 901 20 366 20 774 30 104 30 176.4 30 162 40 33.5 40 62.5 40 52.8 50 15.2 50 33.2 50 35 60 19.5 60 23 222 Appendix 4 Flame spread results 12.1 Summary of flame spread results Plywood - RIFT (using RIFT ignition data) - 17mm RIFT (using ISO ignition data) - 17mm RIFT (using RIFT ignition data) - 17mm RIFT (using ISO 5657 ignition data) - 17mm LIFT - low preheat time low preheat time Full preheat time Full preheat time Full preheat time Flame spread modulus C 7.74 10.72 4.37 4.37 3.7 Flame spread parameter ? 5.80 4.79 18.23 28.84 28.0 " min,ig q (ignition test) 16.25 13.75 16.25 13.75 16.3 " min,ig q (correlation) 16.46 12.83 20.23 20.23 21.6 " s q (correlation) 5.25 5.25 4.62 4.62 5.4 " s q ( from extent of flame spread) 6.47 Std Dev 1.32 6.47 Std Dev1.32 4.74 Std Dev0.33 4.74 Std Dev 0.33 5.3 Std Dev 0.65 Medium Density Fibreboard (MDF) RIFT - 18mm MDF with ISO ign data RIFT - 18mm MDF with RIFT ignition data RIFT - 18mm MDF with ISO ign data RIFT - 18mm MDF with RIFT ignition data LIFT Low preheating time Low preheating time Full preheating time Full preheating time Full preheating time Flame spread modulus C 17.01 17.10 4.9 4.9 4.0 Flame spread parameter ? 1.56 1.55 18.6 18.5 36. " min,ig q (ignition test) 16.25 16.25 16.3 16.3 16.3 " min,ig q (correlation) 14.85 14.82 24.3 24.3 23.6 223 " s q (correlation) 8.67 8.67 6.7 6.7 6.0 " s q ( from extent of flame spread) 8.23 Std Dev 0.26 8.23 Std Dev 0.26 6.3 Std Dev0.2 6.3 Std Dev 0.2 5.9 Std Dev 0.4 Melteca faced Particle board RIFT (using ISO 5657 ignition data RIFT (Using RIFT ign data RIFT (Using RIFT ign data RIFT (Using ISO 5657 ign data LIFT Low preheat time Low preheat time Full preheat time Full preheat time Full preheat time Flame spread modulus C 2.76 2.72 1.78 1.78 Not tested Flame spread parameter ? 90.72 65.54 154.28 206.73 No tested " min,ig q (ignition test) 18.75 23.75 23.75 18.75 Not tested " min,ig q (correlation) 36.60 39.26 62.01 62.01 Not tested " s q (correlation) 11.30 11.30 9.61 9.61 Not tested " s q ( from extent of flame spread) 12.41 Std Dev 1.18 12.41 Std Dev 1.18 11.57 Std Dev1.38 11.57 Std Dev 1.38 Not tested Melteca faced MDF RIFT (using ISO 5657 ignition data RIFT (Using RIFT ign data RIFT (Using RIFT ign data RIFT (Using ISO 5657 ign data LIFT Low preheat time Low preheat time Full preheat time Full preheat time Full preheat time Flame spread modulus C 13.40 2.32 6.44 6.44 4.3 Flame spread parameter ? 3.85 90.36 11.74 16.67 38.5 " min,ig q (ignition test) 20.00 21.25 21.25 20.00 18.8 " min,ig q (correlation) 20.00 46.10 30.74 30.74 33.0 " s q (correlation) 11.30 12.91 11.48 11.48 10.9 " s q ( from extent of flame spread) 14.16 Std Dev 1.63 14.16 Std Dev 1.63 13.96 Std Dev 1.78 13.96 Std Dev 1.78 9.9 Std Dev 0.16 224 Pynefloor Particle board flooring RIFT - with ISO 5657 ign data RIFT - with RIFT ignition data RIFT - with ISO 5657 ignition data RIFT ? with RIFT ignition data LIFT Low preheating time Low preheating time Full preheating time Full preheating time Full preheating time Flame spread modulus C 19.5 9.3 8.1 8.1 3.6 Flame spread parameter ? 2.8 3.7 14.8 4.9 37.3 " min,ig q (ignition test) 0.0 0.0 18.8 21.3 0.0 " min,ig q (correlation) 11.8 20.2 19.0 19.0 25.5 " s q (correlation) 6.6 8.4 7.5 7.5 6.2 " s q ( from extent of flame spread) 8.7 Std Dev 0.74 8.7 Std Dev 0.74 7.6 Std Dev 0.61 7.6 Std Dev 0.61 6.3 Std Dev 0.73 Superflake Particle board RIFT - 20mm particle board with ISO 5657 ign data RIFT - 20mm particle board with RIFT ignition data RIFT - 20mm particle board with ISO 5657 ignition data RIFT - with RIFT ignition data LIFT Low preheating time Low preheating time Full preheating time Full preheating time Full preheating time Flame spread modulus C 24.77 13.79 7.58 7.58 Not Tested Flame spread parameter ? 1.94 2.61 20.71 8.62 Not tested " min,ig q (ignition test) 13.75 18.75 13.75 18.75 Not tested " min,ig q (correlation) 10.44 16.70 18.61 18.61 Not tested " s q (correlation) 6.60 8.39 7.48 7.48 Not tested " s q ( from extent of flame spread) 9.75 Std Dev 1.58 9.75 Std Dev 1.58 7.41 Std Dev 0.71 7.41 Std Dev 0.71 Not tested 225 Hardboard RIFT (using ISO ignition data RIFT (Using RIFT ign data RIFT (using ISO ignition data RIFT (Using RIFT ign data LIFT Low preheat time Low preheat time Full preheat time Full preheat time Full preheat time Flame spread modulus C 13.3 13.6 5.7 5.7 4.9 Flame spread parameter ? 6.7 6.7 36.8 38.1 20.3 " min,ig q (ignition test) 11.3 11.3 11.3 11.3 17.5 " min,ig q (correlation) 11.1 10.9 19.2 19.2 19.5 " s q (correlation) 3.3 3.3 3.3 3.3 3.1 " s q ( from extent of flame spread) 4.1 Std Dev 0.9 4.1 Std Dev 0.9 3.2 Std Dev 0.3 3.2 Std Dev 0.3 3.4 Std Dev 0.4 NZ Beech RIFT (using ISO ignition data RIFT (Using RIFT ign data RIFT (using ISO ignition data RIFT (Using RIFT ign data LIFT Low preheat time Low preheat time Full preheat time Full preheat time Full preheat time Flame spread modulus C 4.7 7.3 2.5 2.5 2.7 Flame spread parameter ? 20.2 20.2 74.8 177.7 70.9 " min,ig q (ignition test) 18.0 18.75 18.00 18.75 18.75 " min,ig q (correlation) 16.2 10.5 22.3 22.3 23.7 " s q (correlation) 6.5 4.2 2.9 2.9 3.7 " s q ( from extent of flame spread) 9.6 Std Dev 2.0 9.6 Std Dev 2.0 3.3 Std Dev 0.3 3.3 Std Dev 0.3 4.8 Std Dev 0.9 226 Radiata Pine RIFT (using ISO ignition data RIFT (Using RIFT ign data RIFT (Using ISO 5657 ign data RIFT (Using RIFT ign data) LIFT Low preheat time Low preheat time Full preheat time Full preheat time Full preheat time Flame spread modulus C 12.3 11.1 5.0 5.0 4.0 Flame spread parameter ? 5.4 5.4 32.6 26.9 27.4 " min,ig q (ignition test) 16.25 18.75 16.25 18.75 18.75 " min,ig q (correlation) 10.2 11.2 17.6 17.6 18.0 " s q (correlation) 4.9 5.3 3.8 3.8 4.2 " s q ( from extent of flame spread) 7.5 Std Dev 1.0 7.5 Std Dev 1.0 3.8 Std Dev 0.1 3.8 Std Dev 0.1 4.5 Std Dev 0.7 Rimu RIFT (using ISO ignition data RIFT (Using RIFT ign data RIFT (using ISO ignition data RIFT (Using RIFT ign data LIFT Low preheat time Low preheat time Full preheat time Full preheat time Full preheat time Flame spread modulus C 4.7 5.1 Insufficient flame spread Insufficient flame spread 2.2 Flame spread parameter ? 22.5 22.5 - - 106 " min,ig q (ignition test) 18.0 21.25 18.0 21.25 18.5 " min,ig q (correlation) 15.9 14.8 Insufficient flame spread Insufficient flame spread 26.0 " s q (correlation) 5.4 5.0 Insufficient flame spread Insufficient flame spread 6.0 " s q ( from extent of flame spread) 11.7 Std Dev 2.3 11.7 std Dev 2.3 12.7 12.7 7.6 Std Dev 2.7 227 Macrocarpa RIFT (using ISO 5657 ignition data) RIFT (Using RIFT ign data) RIFT (Using ISO 5657 ign data) RIFT (Using RIFT ign data LIFT Low preheating time Low preheating time Full preheat time Full preheat time Full preheating time Flame spread modulus C 7.7 6.9 2.9 2.9 2.5 Flame spread parameter ? 9.3 8.6 64.5 48.0 58.3 " min,ig q (ignition test) 18.0 18.8 18.0 18.8 18.8 " min,ig q (correlation) 12.7 14.5 22.6 22.6 18.9 " s q (correlation) 4.1 4.6 4.6 4.6 2.2 . " s q ( from extent of flame spread) 6.1 Std Dev 1.0 6.1 Std Dev 1.0 5.0 Std Dev 0.8 5.0 Std Dev 0.8 2.2 Std Dev 0.2 228 12.2 Flame spread data and correlation results Plywood, 17mm Flame Front Position x Time to position x t (s) time from ignition (s) ?t ?x ?(t 2 ) ?(t) 2 ?(t.x) V f V f F(t) " . . e q )(. " . . tFq e 1/?V f LIFT Plywood Test 1 Ignition Time 295 m/s mm/s kW/m 2 (m/s) -0.5 50 75 100 125 150 175 305 10 1.000 19.66 19.66 200 312 17 939 600 294053 881721 188225 0.0029 2.9110 1.000 19.05 19.05 18.535 225 322 27 967 675 311917 935089 218100 0.0024 2.3792 1.000 18.25 18.25 20.502 250 333 38 996 750 330854 992016 249475 0.0026 2.6099 1.000 17.45 17.45 19.574 275 341 46 1030 825 353906 1060900 283825 0.0021 2.1088 1.000 15.86 15.86 21.776 300 356 61 1073 900 384393 1151329 322775 0.0014 1.4189 1.000 14.28 14.28 26.547 325 376 81 1138 975 432948 1295044 371100 0.0010 0.9868 1.000 12.89 12.89 31.833 350 406 111 1218 1050 496308 1483524 427800 0.0008 0.8333 1.000 11.50 11.50 34.641 375 436 141 1323 1125 586293 1750329 498000 0.0007 0.6579 1.000 10.17 10.17 38.987 400 481 186 1454 1200 709826 2114116 584125 0.0005 0.4931 1.000 8.83 8.83 45.033 425 537 242 1630 1275 894274 2656900 696025 0.0004 0.3790 1.000 7.59 7.59 51.365 450 612 317 1862 1350 1171282 3467044 842300 0.0003 0.2820 1.000 6.35 6.35 59.545 475 713 418 1.000 5.42 5.42 500 870 575 229 Flame Front Position x Time to position x t (s) Time from ignition (s) ?t ?x ?(t 2 ) ?(t) 2 ?(t.x) V f V f F(t) " . . e q )(. " . . tFq e 1/?V f LIFT Plywood Test 2 Ignition Time 307 m/s kW/m 2 (m/s) -0.5 (m/s) -0.5 50 75 100 125 150 175 200 225 319 24 1.000 18.25 18.25 250 328 33 987 750 324945 974169 247275 0.0024 2.3649 1.000 17.45 17.45 20.563 275 340 45 1021 825 347793 1042441 281400 0.0020 1.9989 1.000 15.86 15.86 22.367 300 353 58 1060 900 374898 1123600 318675 0.0019 1.8510 1.000 14.28 14.28 23.243 325 367 72 1108 975 409842 1227664 360975 0.0014 1.4098 1.000 12.89 12.89 26.633 350 388 93 1169 1050 456629 1366561 410325 0.0011 1.0598 1.000 11.50 11.50 30.717 375 414 119 1245 1125 518189 1550025 468250 0.0009 0.9082 1.000 10.17 10.17 33.183 400 443 148 1338 1200 599006 1790244 536875 0.0007 0.7418 1.000 8.83 8.83 36.716 425 481 186 1470 1275 725726 2160900 627325 0.0005 0.4746 1.000 7.59 7.59 45.904 450 546 251 1678 1350 953278 2815684 759350 0.0003 0.2888 1.000 6.35 6.35 58.845 475 651 356 1945 1424 1281421 3783025 928177 0.0002 0.2425 1.000 5.42 5.42 64.214 499 748 453 1.000 4.53 4.53 230 Flame Front Position x Time to position x t (s) Time from ignition (s) ?t ?x ?(t 2 ) ?(t) 2 ?(t.x) V f V f F(t) " . . e q )(. " . . tFq e 1/?V f LIFT Plywood Test 3 Ignition Time 295 m/s kW/m 2 (m/s) -0.5 (m/s) -0.5 50 75 100 125 150 175 200 225 307 12 1.000 18.25 18.25 250 314 19 943 750 296529 889249 236125 0.0033 3.3284 1.000 17.45 17.45 17.333 275 322 27 969 825 313169 938961 266950 0.0026 2.6099 1.000 15.86 15.86 19.574 300 333 38 1001 900 334289 1002001 300900 0.0021 2.0785 1.000 14.28 14.28 21.934 325 346 51 1041 975 361649 1083681 339050 0.0017 1.7180 1.000 12.89 12.89 24.126 350 362 67 1091 1050 397449 1190281 382775 0.0013 1.3432 1.000 11.50 11.50 27.286 375 383 88 1160 1125 449958 1345600 436325 0.0009 0.9300 1.000 10.17 10.17 32.791 400 415 120 1255 1200 527763 1575025 503850 0.0007 0.6716 1.000 8.83 8.83 38.588 425 457 162 1424 1275 685778 2027776 608625 0.0003 0.3476 1.000 7.59 7.59 53.635 450 552 257 1772 1340 1095722 3139984 797420 0.0001 0.1208 1.000 6.35 6.35 90.983 465 763 468 1.000 5.79 5.79 231 Flame Front Position x Time to position x t (s) time from ignition (s) ?t ?x ?(t 2 ) ?(t) 2 ?(t.x) V f F(t) " . . e q )(. " . . tFq e RIFT ? 9mm plywood low preheat time Test 1 Ignition Time 66 m/s kW/m 2 10 20 30 40 50 66 0 0.320 27.20 60 78 12 227 180 17329 51529 13790 0.0011 0.348 24.94 8.68 70 83 17 254 210 21622 64516 17930 0.0013 0.359 22.89 8.22 80 93 27 279 240 26147 77841 22520 0.0010 0.380 20.84 7.92 90 103 37 309 270 32027 95481 28010 0.0010 0.400 18.76 7.50 100 113 47 346 300 40278 119716 34870 0.0007 0.419 16.68 6.99 110 130 64 399 330 54005 159201 44320 0.0005 0.449 15.08 6.77 120 156 90 483 360 80045 233289 58630 0.0003 0.492 13.96 6.87 130 197 131 591 390 119789 349281 77650 0.0002 0.553 12.83 7.10 140 238 172 699 420 165149 488601 98530 0.0003 0.608 11.74 7.14 150 264 198 785 450 206429 616225 118200 0.0004 0.640 10.65 6.81 160 283 217 899 480 273689 808201 144720 0.0002 0.663 9.56 6.33 170 352 286 1357 510 725277 1841449 235080 0.0000 0.739 8.47 6.26 180 72 656 1.000 7.39 7.39 232 Flame Front Position x Time to position x t (s) time from ignition (s) ?t ?x ?(t 2 ) ?(t) 2 ?(t.x) V f F(t) " . . e q )(. " . . tFq e RIFT ? 9mm plywood low preheat time Test 2 Ignition Time 81s m/s kW/m 2 10 20 30 40 81 0 0.355 29.67 50 85 4 257 150 22067 66049 12950 0.0020 0.363 27.20 9.88 60 91 10 271 180 24531 73441 16360 0.0020 0.376 24.94 9.37 70 95 14 285 210 27107 81225 20030 0.0025 0.384 22.89 8.79 80 99 18 299 240 29851 89401 24020 0.0020 0.392 20.84 8.17 90 105 24 318 270 33822 101124 28770 0.0013 0.404 18.76 7.57 100 114 33 346 300 40150 119716 34820 0.0009 0.421 16.68 7.02 110 127 46 386 330 50150 148996 42770 0.0006 0.444 15.08 6.70 120 145 64 441 360 65715 194481 53340 0.0005 0.474 13.96 6.62 130 169 88 512 390 88790 262144 67090 0.0004 0.512 12.83 6.57 140 198 117 589 420 117049 346921 82990 0.0004 0.554 11.74 6.51 150 222 141 671 450 151489 450241 101180 0.0004 0.587 10.65 6.25 160 251 170 0.624 9.56 5.97 Flame Front Position x Time to position x t (s) time from ignition (s) ?t ?x ?(t 2 ) ?(t) 2 ?(t.x) V f F(t) " . . e q )(. " . . tFq e RIFT ? 9mm plywood low preheat time Test 3 Ignition Time 69s m/s kW/m 2 10 20 30 233 40 50 60 69 0 0.327 24.94 70 72 3 216 210 15570 46656 15180 0.0033 0.334 22.89 7.65 80 75 6 229 240 17533 52441 18420 0.0019 0.341 20.84 7.11 90 82 13 252 270 21374 63504 22880 0.0010 0.357 18.76 6.69 100 95 26 287 300 27849 82369 28980 0.0007 0.384 16.68 6.41 110 110 41 339 330 39081 114921 37680 0.0005 0.413 15.08 6.23 120 134 65 400 360 54392 160000 48460 0.0004 0.456 13.96 6.37 130 156 87 471 390 75053 221841 61700 0.0004 0.492 12.83 6.31 140 181 112 540 420 98306 291600 76070 0.0004 0.530 11.74 6.22 150 203 134 603 450 121931 363609 90830 0.0005 0.561 10.65 5.98 160 219 150 666 480 148706 443556 106970 0.0005 0.583 9.56 5.57 170 244 175 742 510 185338 550564 126740 0.0003 0.615 8.47 5.21 180 279 210 842 535 239138 708964 150715 0.0002 0.658 7.39 4.86 185 31 250 0.704 7.02 4.94 Flame Front Position x Time to position x t (s) time from ignition (s) ?t ?x ?(t 2 ) ?(t) 2 ?(t.x) V f F(t) " . . e q )(. " . . tFq e RIFT ? 9mm plywood low preheat time Test 4 Ignition Time 66s m/s kW/m 2 10 20 30 66 0 40 75 9 218 120 15910 47524 8830 0.0016 0.341 29.67 10.12 50 77 11 231 150 17795 53361 11590 0.0050 0.346 27.20 9.40 60 79 13 239 180 19059 57121 14400 0.0032 0.350 24.94 8.73 234 70 83 17 248 210 20526 61504 17430 0.0028 0.359 22.89 8.22 80 86 20 262 240 22934 68644 21060 0.0019 0.365 20.84 7.61 90 93 27 280 270 26246 78400 25350 0.0013 0.380 18.76 7.13 100 101 35 305 300 31171 93025 30680 0.0011 0.396 16.68 6.60 110 111 45 338 330 38398 114244 37430 0.0008 0.415 15.08 6.26 120 126 60 393 360 52533 154449 47610 0.0004 0.442 13.96 6.17 130 156 90 452 390 69112 204304 59200 0.0004 0.492 12.83 6.31 140 170 104 512 420 87832 262144 71980 0.0007 0.514 11.74 6.03 150 186 120 563 450 106345 316969 84820 0.0005 0.537 10.65 5.72 160 207 141 683 480 161545 466489 110320 0.0002 0.567 9.56 5.42 170 290 224 841 510 245285 707281 144340 0.0001 0.671 8.47 5.68 180 344 278 1019 533 350661 ###### 181675 0.0001 0.731 7.39 5.40 183 385 319 0.773 7.17 5.54 Flame Front Position x Time to position x t (s) time from ignition (s) ?t ?x ?(t 2 ) ?(t) 2 ?(t.x) V f F(t) " . . e q )(. " . . tFq e RIFT ? 9mm plywood low preheat time Test 5 Ignition Time 67s m/s kW/m 2 10 20 30 40 50 67 0 140 150 9818 19600 7730 0.0002 0.323 27.20 8.77 60 73 6 221 180 16379 48841 13400 0.0014 0.337 24.94 8.39 70 81 14 245 210 20171 60025 17330 0.0011 0.355 22.89 8.12 80 91 24 276 240 25658 76176 22310 0.0009 0.376 20.84 7.83 90 104 37 313 270 33021 97969 28440 0.0007 0.402 18.76 7.54 235 100 118 51 359 300 43509 128881 36230 0.0006 0.428 16.68 7.14 110 137 70 417 330 58937 173889 46310 0.0005 0.461 15.08 6.95 120 162 95 482 360 78502 232324 58300 0.0004 0.501 13.96 7.00 130 183 116 556 390 104254 309136 72770 0.0004 0.533 12.83 6.84 140 211 144 634 420 135610 401956 89330 0.0004 0.572 11.74 6.72 150 240 173 723 450 176105 522729 109060 0.0003 0.610 10.65 6.50 160 272 205 825 480 229553 680625 132730 0.0003 0.650 9.56 6.21 170 313 246 951 510 305909 904401 162610 0.0002 0.697 8.47 5.91 180 366 299 1102 540 410854 1214404 199460 0.0002 0.754 7.39 5.57 190 423 356 1257 570 531909 1580049 239850 0.0002 0.810 6.66 5.40 200 468 401 1423 600 680977 2024929 285690 0.0002 0.852 5.93 5.06 210 532 465 1600 630 862048 2560000 337320 0.0002 0.909 5.25 4.77 220 600 533 1805 660 1095953 3258025 398510 0.0001 0.965 4.62 4.46 230 673 606 2003 690 1345829 4012009 461990 0.0002 1.022 3.99 4.08 240 730 663 2246 715 1696478 5044516 536525 0.0001 1.000 3.74 3.74 245 843 776 1.000 3.62 3.62 Flame Front Position x Time to position x t (s) Time from ignition (s) ?t ?x ?(t 2 ) ?(t) 2 ?(t.x) V f F(t) " . . e q )(. " . . tFq e RIFT ? 9mm plywood low preheat time Test 6 Ignition Time 66s m/s kW/m 2 10 20 30 40 50 60 67 0 0.0002 0.323 24.94 70 78 11 239 210 19409 57121 17000 0.0007 0.348 22.89 7.96 236 80 94 27 281 240 26801 78961 22790 0.0006 0.382 20.84 7.96 90 109 42 330 270 36846 108900 30030 0.0006 0.411 18.76 7.72 100 127 60 381 300 49035 145161 38460 0.0006 0.444 16.68 7.41 110 145 78 435 330 63723 189225 48210 0.0006 0.474 15.08 7.15 120 163 96 503 360 85619 253009 60860 0.0004 0.503 13.96 7.02 130 195 128 586 390 116578 343396 76830 0.0003 0.550 12.83 7.06 140 228 161 713 419 174109 508369 100480 0.0002 0.595 11.74 6.98 149 290 223 0.671 10.75 7.22 237 Flame Front Position x Time to position x t (s) Time from ignition (s) ?t ?x ?(t 2 ) ?(t) 2 ?(t.x) V f F(t) " . . e q )(. " . . tFq e RIFT ? 17mm plywood low preheat time Test 1 Ignition Time 69 m/s kW/m 2 10 69 0 37.09 20 69.5 0.5 208 60 14422 43264 4165 0.0300 0.401 34.62 13.89 30 9.5 0. 209 90 14561 43681 6275 0.0300 0.401 32.14 12.8 40 70 1 213 120 15059 45156 8535 0.0049 0.403 29.67 11.94 50 73 4 219 150 16005 47961 11010 0.0033 0.411 27.20 11.18 60 76 7 234 180 18330 54756 14160 0.0015 0.419 24.94 10.46 70 85 16 253 210 21465 64009 17870 0.0012 0.444 22.89 10.15 80 92 23 280 240 26298 78400 22580 0.0011 0.461 20.84 9.62 90 103 34 310 270 32298 96100 28130 0.0009 0.488 18.76 9.16 100 115 46 350 300 41258 122500 3529 0.0007 0.516 16.68 8.61 110 132 63 397 330 53149 157609 44020 0.0006 0.553 15.08 8.34 120 150 81 454 360 69508 206116 5488 0.0005 0.589 13.96 8.22 130 172 103 520 390 91288 270400 68080 0.0004 0.631 12.83 8.10 140 198 129 600 420 121688 36000 8458 0.0003 0.677 11.74 7.95 150 230 161 704 450 168280 495616 106380 0.0003 0.730 10.65 7.77 160 276 207 815 480 224557 664225 131190 0.0003 0.799 9.56 7.64 170 309 240 946 510 301978 894916 161670 0.0002 0.846 8.47 7.17 180 361 292 1082 540 395546 1170724 195790 0.0002 0.914 7.39 6.75 190 412 343 1194 570 477306 1425636 227460 0.0003 0.977 6.66 6.50 200 421 352 1401 598 669609 1962801 280624 0.0001 0.987 5.93 5.85 208 568 499 5.39 238 Flame Front Position x Time to position x t (s) Time from ignition (s) ?t ?x ?(t 2 ) ?(t) 2 ?(t.x) V f F(t) " . . e q )(. " . . tFq e RIFT ? 17mm plywood low preheat time Test 2 Ignition Time 50 m/s kW/m 2 10 50 0 0.340 37.09 20 52 2 155 60 8013 24025 3130 0.0064 0.347 34.62 12.01 30 53 3 159 90 8429 25281 4790 0.0100 0.350 32.14 11.26 40 54 4 162 120 8750 26244 6500 0.0100 0.354 29.67 10.49 50 55 5 169 150 9541 28561 8510 0.0029 0.357 27.20 9.70 60 60 10 181 180 10981 32761 10970 0.0018 0.373 24.94 9.29 70 66 16 201 210 13581 40401 14220 0.0013 0.391 22.89 8.95 80 75 25 229 240 17725 52441 18540 0.0009 0.417 20.84 8.68 90 88 38 264 270 23570 69696 24020 0.0008 0.451 18.76 8.47 100 101 51 305 300 31401 93025 30780 0.0007 0.484 16.68 8.06 110 116 66 349 330 41081 121801 3870 0.0006 0.518 15.08 7.81 120 132 82 400 360 53984 160000 48360 0.0006 0.553 13.96 7.71 130 152 102 461 390 71857 212521 6038 0.0004 0.593 12.83 7.61 140 177 127 536 420 97282 287296 75590 0.0004 0.640 11.74 7.51 150 207 157 626 450 132742 39187 9455 0.0003 0.692 10.65 7.37 160 242 192 732 480 181502 535824 117880 0.0003 0.748 9.56 7.15 170 283 233 0.809 8.47 Flame Front Position x Time to position x t (s) Time from ignition (s) ?t ?x ?(t 2 ) ?(t) 2 ?(t.x) V f F(t) " . . e q )(. " . . tFq e RIFT ? 17mm plywood low preheat time Test 3 Ignition Time 69 m/s kW/m 2 10 50 0 0.340 37.09 20 52 2 155 60 8013 24025 3130 0.0064 0.347 34.62 12.01 30 53 3 159 90 8429 25281 4790 0.0100 0.350 32.14 11.26 40 54 4 162 120 8750 26244 6500 0.0100 0.354 29.67 10.49 50 55 5 169 150 9541 28561 8510 0.0029 0.357 27.20 9.70 239 60 60 10 181 180 10981 32761 10970 0.0018 0.373 24.94 9.29 70 66 16 201 210 13581 40401 14220 0.0013 0.391 22.89 8.95 80 75 25 229 240 17725 52441 18540 0.0009 0.417 20.84 8.68 90 88 38 264 270 23570 69696 24020 0.0008 0.451 18.76 8.47 100 101 51 305 300 31401 93025 30780 0.0007 0.484 16.68 8.06 110 116 66 349 330 41081 121801 3870 0.0006 0.518 15.08 7.81 120 132 82 400 360 53984 160000 48360 0.0006 0.553 13.96 7.71 130 152 102 461 390 71857 212521 6038 0.0004 0.593 12.83 7.61 140 177 127 536 420 97282 287296 75590 0.0004 0.640 11.74 7.51 150 207 157 626 450 132742 391876 94550 0.0003 0.692 10.65 7.37 160 242 192 732 480 181502 535824 117880 0.0003 0.748 9.56 7.15 170 283 233 859 510 250209 737881 146950 0.0002 0.809 8.47 6.86 180 334 284 999 539 337569 998001 180428 0.0002 0.879 7.39 6.50 189 382 332 0.940 6.73 Flame Front Position x Time to position x t (s) Time from ignition (s) ?t ?x ?(t 2 ) ?(t) 2 ?(t.x) V f F(t) " . . e q )(. " . . tFq e RIFT ? 17mm plywood low preheat time Test 4 Ignition Time 69 m/s kW/m 2 10 50 0 0.340 37.09 20 52 2 156 60 8120 24336 3160 0.0050 0.347 34.62 12.01 30 54 4 164 90 8984 2689 4980 0.0032 0.354 32.14 11.36 40 58 8 173 120 10001 29929 6990 0.0028 0.366 29.67 10.87 50 61 11 186 150 11574 34596 9390 0.0021 0.376 27.20 10.22 60 67 17 204 180 13986 41616 12390 0.0013 0.394 24.94 9.82 70 76 26 229 210 17661 52441 16220 0.0011 0.419 22.89 9.60 80 86 36 259 240 22581 67081 20930 0.0010 0.446 20.84 9.30 90 97 47 297 270 29801 88209 27010 0.0007 0.474 18.76 8.89 100 114 64 343 300 39829 117649 3465 0.0006 0.514 16.68 8.57 240 110 132 82 395 330 52621 156025 43800 0.0006 0.553 15.08 8.34 120 149 99 456 360 70250 207936 5515 0.0005 0.587 13.96 8.20 130 175 125 531 390 95675 281961 69610 0.0003 0.636 12.83 8.17 140 207 157 615 420 127763 378225 86680 0.0003 0.692 11.74 8.13 150 233 183 708 450 168962 501264 106810 0.0003 0.734 10.65 7.82 160 268 218 807 480 219749 651249 129850 0.0003 0.788 9.56 7.53 170 306 256 901 510 272389 811801 153760 0.0003 0.842 8.47 7.13 180 327 277 0.870 7.39 Flame Front Position x Time to position x t (s) Time from ignition (s) ?t ?x ?(t 2 ) ?(t) 2 ?(t.x) V f F(t) " . . e q )(. " . . tFq e RIFT ? 17mm plywood low preheat time Test 5 Ignition Time 69 m/s kW/m 2 10 50 0 0.340 37.09 20 52.5 2.5 158 60 8281 24806 3200 0.0040 0.349 34.62 12.07 30 55 168 90 938 2805 5100 0.0026 0.357 32.14 11.4 40 60 10 184 120 11386 33856 7500 0.0014 0.373 29.67 11.06 50 69 19 205 150 14137 42025 10410 0.0012 0.400 27.20 10.87 60 76 26 227 180 17261 51529 13750 0.0015 0.419 24.94 10.46 70 82 32 247 210 20421 61009 17420 0.0015 0.436 22.89 9.97 80 89 39 271 240 24645 73441 21860 0.0011 0.454 20.84 9.46 90 100 50 304 270 31146 92416 27620 0.0008 0.481 18.76 9.03 100 115 65 352 300 41994 123904 3557 0.0005 0.516 16.68 8.61 110 137 87 406 330 55710 164836 45050 0.0005 0.563 15.08 8.49 120 154 104 469 360 74169 219961 5669 0.0005 0.597 13.96 8.33 130 178 128 545 390 100769 297025 71440 0.0003 0.642 12.83 8.24 140 213 163 638 420 138062 407044 90010 0.0003 0.702 11.74 8.24 241 150 247 197 750 450 190478 562500 113270 0.0003 0.756 10.65 8.05 160 290 240 880 480 262758 774400 141760 0.0002 0.819 9.56 7.83 170 343 293 1027 510 356985 1054729 175630 0.0002 0.891 8.47 7.55 180 394 344 1188 540 476286 1411344 214920 0.0002 0.955 7.39 7.06 190 451 401 1357 570 620781 1841449 259010 0.0002 1.000 6.66 6.66 200 512 462 1.000 5.93 Flame Front Position x Time to position x t (s) Time from ignition (s) ?t ?x ?(t 2 ) ?(t) 2 ?(t.x) V f F(t) " . . e q )(. " . . tFq e RIFT ? 17mm plywood low preheat time Test 6 Ignition Time 69 m/s kW/m 2 10 51 0 0.344 37.09 20 52 1 158 60 8330 24964 3200 0.0046 0.347 34.62 12.01 30 55 4 167 90 9329 27889 5090 0.0024 0.357 32.14 11.47 40 60 9 182 120 11114 33124 7400 0.0017 0.373 29.67 11.06 50 67 16 204 150 14018 41616 10370 0.0012 0.394 27.20 10.71 60 77 26 234 180 18518 54756 14270 0.0009 0.422 24.94 10.53 70 90 39 267 210 24029 71289 18920 0.0009 0.456 22.89 10.45 80 100 49 305 240 31325 93025 24650 0.0008 0.481 20.84 10.03 90 115 64 345 270 40125 119025 3135 0.0007 0.516 18.76 9.68 100 130 79 397 300 53229 157609 40070 0.0005 0.549 16.68 9.15 110 152 101 455 330 69933 207025 5048 0.0005 0.593 15.08 8.94 120 173 122 531 360 95469 281961 64260 0.0004 0.633 13.96 8.83 130 206 155 615 390 128061 378225 8058 0.0003 0.691 12.83 8.86 140 236 185 715 420 172661 511225 100770 0.0003 0.739 11.74 8.68 150 273 222 818 450 225706 669124 123430 0.0003 0.795 10.65 8.46 160 309 258 934 480 293914 872356 150230 0.0003 0.846 9.56 8.08 242 170 352 301 1065 510 382601 1134225 182000 0.0002 0.903 8.47 7.65 180 404 353 1216 540 498720 1478656 219960 0.0002 0.967 7.39 7.15 190 460 409 1378 570 639012 1898884 262920 0.0002 1.000 6.66 6.66 200 514 463 1550 600 807572 2402500 311160 0.0002 1.000 5.93 5.93 210 576 525 1710 625 980372 2924100 357060 0.0001 1.000 5.25 5.25 215 620 569 1.000 4.94 Flame Front Position x Time to position x t (s) Time from ignition (s) ?t ?x ?(t 2 ) ?(t) 2 ?(t.x) V f F(t) " . . e q )(. " . . tFq e RIFT ? 17mm plywood full preheat time Test 1 Ignition Time 291 m/s kW/m 2 10 0 0 20 0 0 30 0 0 40 0 0 50 0 0 60 0 0 70 0 0 80 0 0 90 296 5 18.76 100 301 10 906 300 273698 820836 90730 0.0015 1.000 16.68 16.68 110 309 18 932 330 289766 868624 102730 0.0009 1.000 15.08 15.08 120 322 31 964 360 310054 929296 115920 0.0008 1.000 13.96 13.96 130 333 42 1000 390 333598 1000000 130230 0.0009 1.000 12.83 12.83 140 345 54 1071 420 384363 1147041 150540 0.0003 1.000 11.74 11.74 150 393 102 1155 450 447363 1334025 173970 0.0003 1.000 10.65 10.65 160 417 126 1249 480 521059 1560001 200300 0.0004 1.000 9.56 9.56 170 439 148 1323 510 584699 1750329 225410 0.0004 1.000 8.47 8.47 180 467 176 1399 540 653859 1957201 252360 0.0004 1.000 7.39 7.39 190 493 202 1498 570 750582 2244004 285330 0.0003 1.000 6.66 6.66 200 538 247 1624 598 884142 2637376 324614 0.0002 1.000 5.93 5.93 243 208 593 302 1.000 5.39 Flame Front Position x Time to position x t (s) Time from ignition (s) ?t ?x ?(t 2 ) ?(t) 2 ?(t.x) V f F(t) " . . e q )(. " . . tFq e RIFT ? 17mm plywood full preheat time Test 2 Ignition Time 329 m/s kW/m 2 10 0 0 2 30 4 50 6 70 8 90 100 337 8 1.000 16.68 110 340 11 1026 330 350970 1052676 112980 0.0015 1.000 15.08 15.08 120 349 20 1048 360 366282 1098304 125950 0.0011 1.000 13.96 13.96 130 359 30 1079 390 388323 1164241 140490 0.0009 1.000 12.83 12.83 140 371 42 1115 420 414747 1243225 156360 0.0008 1.000 11.74 11.74 150 385 56 1159 450 448275 1343281 174170 0.0006 1.000 10.65 10.65 160 403 74 1276 480 548778 1628176 205190 0.0002 1.000 9.56 9.56 170 488 159 1408 510 667842 1982464 240500 0.0002 1.000 8.47 8.47 180 517 188 1497 540 747497 2241009 269500 0.0001 1.000 7.39 7.39 190 492 163 1538 570 789194 2365444 292340 0.0002 1.000 6.66 6.66 200 529 200 1588 600 843394 2521744 318350 0.0003 1.000 5.93 5.93 210 567 238 1699 630 964939 2886601 357530 0.0003 1.000 5.25 5.25 220 603 274 1796 653 1076974 3225616 391328 0.0002 1.000 4.62 4.62 244 223 626 297 1.000 4.43 Flame Front Position x Time to position x t (s) Time from ignition (s) ?t ?x ?(t 2 ) ?(t) 2 ?(t.x) V f F(t) " . . e q )(. " . . tFq e RIFT ? 17mm plywood full preheat time Test 3 Ignition Time 432 m/s kW/m 2 10 2 30 4 50 60 7 80 9 100 435 3 1.000 16.68 110 442 10 1328 330 587990 1763584 146240 0.0012 1.000 15.08 15.08 120 451 19 1358 360 614990 1844164 163190 0.0009 1.000 13.96 13.96 130 465 33 1397 390 650987 1951609 181910 0.0007 1.000 12.83 12.83 140 481 49 1449 420 700595 2099601 203240 0.0005 1.000 11.74 11.74 150 503 71 1511 450 762099 2283121 227110 0.0004 1.000 10.65 10.65 160 527 95 1584 480 837654 2509056 253950 0.0004 1.000 9.56 9.56 170 554 122 1673 510 935109 2798929 285060 0.0003 1.000 8.47 8.47 180 592 160 1795 540 1078581 3222025 324050 0.0002 1.000 7.39 7.39 190 649 217 1955 570 1281461 3822025 372670 0.0002 1.000 6.66 6.66 200 714 282 2126 600 1513166 4519876 426340 0.0002 1.000 5.93 5.93 210 763 331 2306 630 1779206 5317636 485410 0.0002 1.000 5.25 5.25 245 220 829 397 1.000 4.62 Flame Front Position x Time to position x t (s) Time from ignition (s) ?t ?x ?(t 2 ) ?(t) 2 ?(t.x) V f F(t) " . . e q )(. " . . tFq e RIFT ? 17mm plywood full preheat time Test 4 Ignition Time 455 m/s kW/m 2 10 0 0 20 0 0 30 0 0 40 0 0 50 0 0 60 0 0 70 0 0 80 0 0 90 0 0 100 462 7 16.68 110 467 12 1402 330 655262 1965604 154330 0.0018 1.000 15.08 15.08 120 473 18 1420 360 672218 2016400 170530 0.0015 1.000 13.96 13.96 130 480 25 1442 390 693250 2079364 187620 0.0012 1.000 12.83 12.83 140 489 34 1469 420 719521 2157961 205860 0.0010 1.000 11.74 11.74 150 500 45 1505 450 755377 2265025 226020 0.0007 1.000 10.65 10.65 160 516 61 1550 480 801412 2402500 248340 0.0006 1.000 9.56 9.56 170 534 79 1609 510 863893 2588881 273960 0.0005 1.000 8.47 8.47 180 559 104 1678 540 939862 2815684 302550 0.0004 1.000 7.39 7.39 190 585 130 1766 570 1041590 3118756 336170 0.0003 1.000 6.66 6.66 200 622 167 1864 600 1160758 3474496 373520 0.0003 1.000 5.93 5.93 210 657 202 1961 630 1283657 3845521 412410 0.0003 1.000 5.25 5.25 220 682 227 1.000 4.62 246 Flame Front Position x Time to position x t (s) Time from ignition (s) ?t ?x ?(t 2 ) ?(t) 2 ?(t.x) V f F(t) " . . e q )(. " . . tFq e RIFT ? 17mm plywood full preheat time Test 5 Ignition Time 432 m/s kW/m 2 10 0 0 20 0 0 30 0 0 40 0 0 50 0 0 60 0 0 70 0 0 80 0 0 90 0 0 100 439 7 16.68 110 443 11 1331 330 590571 1771561 146510 0.0020 1.000 15.08 15.08 120 449 17 1350 360 607614 1822500 162150 0.0013 1.000 13.96 13.96 130 458 26 1375 390 630389 1890625 178940 0.0011 1.000 12.83 12.83 140 468 36 1405 420 658229 1974025 196910 0.0010 1.000 11.74 11.74 150 479 47 1441 450 692501 2076481 216410 0.0008 1.000 10.65 10.65 160 494 62 1483 480 733577 2199289 237590 0.0006 1.000 9.56 9.56 170 510 78 1537 510 788225 2362369 261680 0.0005 1.000 8.47 8.47 180 533 101 1611 540 866813 2595321 290560 0.0003 1.000 7.39 7.39 190 568 136 1714 570 982482 2937796 326460 0.0002 1.000 6.66 6.66 200 613 181 1842 600 1135314 3392964 369330 0.0002 1.000 5.93 5.93 210 661 229 2006 629 1348514 4024036 421718 0.0002 1.000 5.25 5.25 219 732 300 1.000 4.68 247 Flame Front Position x Time to position x t (s) Time from ignition (s) ?t ?x ?(t 2 ) ?(t) 2 ?(t.x) V f F(t) " . . e q )(. " . . tFq e RIFT ? 17mm plywood full preheat time Test 6 Ignition Time 432 m/s kW/m 2 10 0 0 20 0 0 30 0 0 40 0 0 50 0 0 60 0 0 70 0 0 80 0 0 90 0 0 100 439 7 1.000 16.68 110 443 11 1331 330 590571 1771561 146510 0.0020 1.000 15.08 15.08 120 449 17 1350 360 607614 1822500 162150 0.0013 1.000 13.96 13.96 130 458 26 1375 390 630389 1890625 178940 0.0011 1.000 12.83 12.83 140 468 36 1405 420 658229 1974025 196910 0.0010 1.000 11.74 11.74 150 479 47 1441 450 692501 2076481 216410 0.0008 1.000 10.65 10.65 160 494 62 1483 480 733577 2199289 237590 0.0006 1.000 9.56 9.56 170 510 78 1537 510 788225 2362369 261680 0.0005 1.000 8.47 8.47 180 533 101 1611 540 866813 2595321 290560 0.0003 1.000 7.39 7.39 190 568 136 1714 570 982482 2937796 326460 0.0002 1.000 6.66 6.66 200 613 181 1842 600 1135314 3392964 369330 0.0002 1.000 5.93 5.93 210 661 229 2006 629 1348514 4024036 421718 0.0002 1.000 5.25 5.25 219 732 300 1.000 4.68 248 Medium Density Fibreboard (MDF) Flame Front Position x Time to position x t (s) Time from ignition (s) ?t ?x ?(t 2 ) ?(t) 2 ?(t.x) V f F(t) " . . e q )(. " . . tFq e 1/?V f LIFT ? 18mm MDF Test 1 Ignition Time 479 m/s kW/m 2 (m/s) -0.5 50 479 7 10 479 25 10 479 75 20 479 225 482 3 18.54 250 488 9 1462 750 712532 2137444 365750 0.0049 1.000 17.25 17.25 14.236 275 492 13 1510 825 761108 2280100 416300 0.0010 1.000 15.85 15.85 31.992 300 530 51 1598 900 854740 2553604 481500 0.0006 1.000 14.45 14.45 41.050 325 576 97 1746 975 1022276 3048516 570200 0.0005 1.000 12.84 12.84 47.113 350 640 161 1924 1050 1242640 3701776 67670 0.0004 1.00 11.23 11.23 51.389 375 708 229 2159 1125 1568585 4661281 813900 0.0003 1.000 9.64 9.64 58.888 400 811 332 2461 1200 2046349 6056521 990250 0.0002 1.000 8.05 8.05 68.574 425 942 463 2887 1275 2831041 8334769 1235050 0.0002 1.000 7.01 7.01 80.850 450 1134 655 3308 1335 3691144 10942864 1477370 0.0001 1.000 5.97 5.97 90.534 460 1232 753 1.000 5.56 5.56 249 Flame Front Position x Time to position x t (s) Time from ignition (s) ?t ?x ?(t 2 ) ?(t) 2 ?(t.x) V f F(t) " . . e q )(. " . . tFq e 1/?V f x LIFT ? 18mm MDF Test 2 Ignition Time 478 m/s kW/m 2 (m/s) -0.5 50 478 7 10 478 25 10 478 75 20 478 5 20 478 275 487 9 15.85 300 515 37 1558 900 811530 2427364 469125 0.0007 1.000 14.45 14.45 37.367 325 556 78 1678 975 942810 2815684 547650 0.0005 1.000 12.84 12.84 42.980 350 607 129 1850 1050 1149554 3422500 650775 0.0004 1.00 11.23 11.23 51.602 375 687 209 2098 1125 1486834 4401604 791675 0.0003 1.000 9.64 9.64 63.137 400 804 326 2418 1200 1977714 5846724 973200 0.0002 1.000 8.05 8.05 69.289 425 927 449 2820 1266 1712770 7952400 738325 0.0005 1.000 7.01 7.01 45.570 441 1089 611 1.000 6.35 6.35 Flame Front Position x Time to position t time from ignition ?t ?x ?(t 2 ) ?(t) 2 ?(t.x) V F(t) q? q?.F(t) 1/?V LIFT ? 18mm MDF Test 3 Ignition Time 455s m/s kW/m 2 (m/s) -0.5 50 45 7 10 45 250 125 45 0 175 45 20 5 45 20 275 466 11 15.85 300 487 32 1478 900 729950 2184484 444875 0.0008 1.000 14.45 14.45 34.823 325 525 70 1589 975 845723 2524921 518675 0.0006 1.000 12.84 12.84 42.597 350 577 122 1740 1050 1015598 3027600 61182 0.0004 1.00 11.23 11.23 47.590 375 638 183 1950 1125 1280198 3802500 735200 0.0003 1.000 9.64 9.64 56.698 400 735 280 2232 1200 1685150 4981824 898325 0.0002 1.000 8.05 8.05 66.648 425 859 404 2618 1275 1278106 6853924 659075 0.0005 1.000 7.01 7.01 47.107 450 1024 569 2992 1332 1786457 8952064 825875 0.0004 1.000 5.97 5.97 48.815 457 1109 654 1.000 5.69 5.69 Flame Front Position x Time to position x t (s) time from ignition ?t ?x ?(t 2 ) ?(t) 2 ?(t.x) V f F(t) " . . e q )(. " . . tFq e 1/?V f MDF-RIFT - low preheat ISO ign data test 1 Ignition Time 69s m/s kW/m 2 (m/s) -0.5 10 69 0 34.0 20 207 60 14283 42849 4140 #DIV/0! 0.441 33.1 14.57 30 69 0 207 90 14283 4284 6210 #DIV/0 0.441 32.5 14.31 40 211 120 14851 44521 8480 0.0038 0.441 30.0 13.22 16.330 50 73 4 226 150 17146 51076 11450 0.0012 0.453 27.5 12.48 28.363 60 84 15 256 180 22186 6553 15620 0.0008 0.486 25.3 12.31 36.197 70 99 30 301 210 30781 90601 21410 0.0006 0.528 23.4 12.33 41.326 80 118 49 356 240 43046 126736 28880 0.0005 0.576 21.4 12.33 44.740 90 139 70 428 270 62486 183184 39050 0.0004 0.625 19.2 11.99 51.846 100 171 102 516 300 90998 266256 52270 0.0003 0.694 16.9 11.75 57.899 110 206 137 626 330 133678 39187 69640 0.0003 0.761 15.0 11.45 62.559 251 120 249 180 751 360 192053 564001 91020 0.0002 0.837 13.5 11.29 67.104 130 296 227 893 390 270721 797449 117080 0.0002 0.913 11.9 10.90 70.386 140 348 279 1052 420 375184 11067604 148400 0.0002 0.990 10.8 10.72 74.897 150 408 339 1243 450 524737 1545049 187840 0.0001 1.000 9.7 9.72 83.626 160 487 418 1.000 8.7 8.67 Flame Front Position x Time to position x t (s) time from ignition ?t ?x ?(t 2 ) ?(t) 2 ?(t.x) V f F(t) " . . e q )(. " . . tFq e 1/?V f MDF-RIFT - low preheat ISO ign data test 2 Ignition Time 60s m/s kW/m 2 (m/s) -0.5 34.9 10 60 0 0.411 34.0 13.96 20 61 1 184 60 11290 33856 3710 0.0064 0.414 33.1 13.70 12.472 30 63 3 190 90 12046 36100 5750 0.0039 0.421 32.5 13.67 15.916 40 66 6 203 120 13801 41209 8230 0.0017 0.431 30.0 12.93 24.246 50 74 14 227 150 17401 5152 11560 0.0009 0.456 27.5 12.57 32.708 60 87 27 263 180 23449 69169 16060 0.0007 0.495 25.3 12.53 37.448 70 102 42 306 210 31662 93636 21720 0.0007 0.536 23.4 12.52 38.730 80 117 57 355 240 42589 126025 28740 0.0006 0.574 21.4 12.28 41.326 90 136 76 422 270 60746 178084 38500 0.0004 0.619 19.2 11.86 51.603 100 169 109 498 300 84306 24800 50370 0.0003 0.690 16.9 11.68 53.607 110 193 133 594 330 119634 352836 65970 0.0003 0.737 15.0 11.08 56.653 120 232 172 704 360 168914 49561 85340 0.0002 0.808 13.5 10.90 65.669 130 279 219 842 390 241226 708964 110450 0.0002 0.886 11.9 10.58 70.386 140 331 271 999 42 338723 998001 140960 0.0002 0.965 10.8 10.45 74.199 150 89 329 1183 450 475251 1399489 178770 0.0002 1.000 9.7 9.72 81.439 160 463 403 1341 32 369171 1798281 133020 0.0000 1.00 8.7 8.67 151.590 164 489 429 1.000 8.1 8.08 252 Flame Front Position x Time to position x t (s) time from ignition ?t ?x ?(t 2 ) ?(t) 2 ?(t.x) V f F(t) " . . e q )(. " . . tFq e 1/?V f MDF-RIFT - low preheat ISO ign data test 3 Ignition Time 59s m/s kW/m 2 (m/s) -0.5 10 59 0 0.407 34.0 20 177 60 10443 31329 3540 #DIV/0! 0.407 33.1 13.47 ! 30 59 0 183 90 11187 3348 5550 0.0025 0.407 32.5 13.23 20.000 40 65 6 199 120 13331 39601 8120 0.0012 0.428 30.0 12.83 28.577 50 75 16 228 150 17594 51984 11630 0.0009 0.459 27.5 12.65 34.008 60 88 29 266 180 23978 70756 16240 0.0007 0.498 25.3 12.60 37.448 70 103 44 310 210 32514 96100 22010 0.0006 0.538 23.4 12.58 39.377 80 119 60 360 240 43814 12960 29150 0.0006 0.579 21.4 12.39 41.884 90 138 79 422 270 60430 178084 38440 0.0004 0.623 19.2 11.94 48.199 100 165 106 499 300 84685 249001 50480 0.0003 0.681 16.9 11.54 53.894 110 196 137 598 330 121810 357604 66500 0.0003 0.743 15.0 11.17 60.193 120 237 178 713 360 172985 508369 86400 0.0002 0.817 13.5 11.02 64.814 130 280 221 847 390 243469 717409 111040 0.0002 0.888 11.9 10.60 68.255 140 330 271 1001 42 340181 1002001 141250 0.0002 0.964 10.8 10.44 74.620 150 91 332 1199 450 490265 1437601 181330 0.0001 1.000 9.7 9.72 86.465 160 478 419 1397 37 388761 1951609 140290 0.0001 1.00 8.7 8.67 90.437 164 528 469 1.000 8.1 8.08 Flame Front Position x Time to position t time from ignition ?t ?x ?(t 2 ) ?(t) 2 ?(t.x) V F(t) q? q?.F(t) 1/?V MDF-RIFT - low preheat ISO ign data test 4 Ignition Time 66s m/s kW/m 2 (m/s) -0.5 10 66 0 0.431 34.0 20 67 1 60 13469 0 4040 0.0003 0.434 33.1 14.36 57.740 30 68 2 90 14013 6180 0.0004 0.437 32.5 14.20 47.618 40 70 4 212 120 15000 44944 8540 0.0032 0.444 30.0 13.32 17.638 50 74 8 230 150 17772 52900 11660 0.0012 0.456 27.5 12.57 29.439 253 60 86 20 260 180 22872 67600 15860 0.0008 0.492 25.3 12.46 36.091 70 100 34 304 210 31320 92416 21600 0.0006 0.530 23.4 12.39 40.104 80 118 52 355 240 42693 126025 28770 0.0005 0.576 21.4 12.33 43.017 90 137 71 422 270 60582 178084 38470 0.0004 0.621 19.2 11.90 49.911 100 16 101 502 300 85862 25200 50810 0.0003 0.686 16.9 11.61 55.229 110 198 132 602 330 123262 362404 66920 0.0003 0.746 15.0 11.23 59.289 120 237 171 714 360 173214 509796 86490 0.0002 0.817 13.5 11.02 63.654 130 279 213 853 390 247579 727609 111890 0.0002 0.886 11.9 10.58 71.012 140 337 271 1010 42 346646 1020100 142550 0.0002 0.974 10.8 10.55 75.830 150 94 328 1257 450 545481 1580049 190440 0.0001 1.000 9.7 9.72 99.730 160 526 460 1460 473 723512 2131600 231280 0.0001 1.00 8.7 8.67 109.287 163 540 474 1.000 8.2 8.23 Flame Front Position x Time to position x t (s) time from ignition ?t ?x ?(t 2 ) ?(t) 2 ?(t.x) V f F(t) " . . e q )(. " . . tFq e 1/?V f MDF-RIFT - low preheat ISO ign data test 5 Ignition Time 61s m/s kW/m 2 (m/s) -0.5 10 61 0 0.414 34.0 14.08 20 62 1 187 60 11661 34969 3770 0.0064 0.418 33.1 13.81 12.472 30 64 3 195 90 12701 38025 5920 0.0027 0.424 32.5 13.78 19.272 40 69 8 207 120 14333 42849 8380 0.0020 0.441 30.0 13.22 22.361 50 74 13 229 150 17633 52441 11620 0.0011 0.456 27.5 12.57 29.967 60 86 25 259 180 22673 6708 15790 0.0008 0.492 25.3 12.46 35.365 70 99 38 305 210 31597 93025 21690 0.0006 0.528 23.4 12.33 41.610 80 120 59 360 240 44082 129600 29220 0.0005 0.581 21.4 12.44 45.826 90 141 80 432 270 63522 186624 39390 0.0004 0.630 19.2 12.07 50.759 100 17 110 513 300 89523 263169 51900 0.0003 0.694 16.9 11.75 54.772 110 201 140 614 330 128206 376996 68250 0.0003 0.752 15.0 11.31 59.820 120 242 181 731 360 181909 534361 88590 0.0002 0.825 13.5 11.13 65.991 130 288 227 875 390 260533 765625 114780 0.0002 0.900 11.9 10.75 71.900 140 345 284 1037 42 365185 1075369 146340 0.0002 0.985 10.8 10.67 76.162 254 150 404 343 1230 450 513602 1512900 185860 0.0001 1.000 9.7 9.72 82.703 160 81 420 1439 472 701493 2070721 227308 0.0001 1.00 8.7 8.67 111.487 162 554 493 1.000 8.4 8.37 Flame Front Position x Time to position x t (s) time from ignition ?t ?x ?(t 2 ) ?(t) 2 ?(t.x) V f F(t) " . . e q )(. " . . tFq e 1/?V f MDF-RIFT - low preheat. ISO ign data test 6 Ignition Time 68s m/s kW/m 2 (m/s) -0.5 10 68 0 0.437 34.0 14.86 20 68.5 0.5 206 60 14077 42230 4120 0.0200 0.439 33.1 14.52 7.071 30 69 1 208 90 14353 43056 6240 0.0129 0.441 32.5 14.31 8.819 40 70 2 215 120 15437 46225 8670 0.0024 0.444 30.0 13.32 20.237 50 76 8 237 150 18957 56169 12060 0.0009 0.462 27.5 12.74 33.381 60 91 23 271 180 24873 73441 1654 0.0007 0.506 25.3 12.82 37.448 70 104 36 315 210 33497 99225 22340 0.0007 0.541 23.4 12.64 38.147 80 120 52 366 240 45380 133956 29660 0.0005 0.581 21.4 12.44 43.770 90 142 74 433 270 63805 187489 39480 0.0004 0.632 19.2 12.12 50.656 100 171 103 515 300 90209 265225 52100 0.0003 0.694 16.9 11.75 54.782 110 202 134 614 330 128126 376996 6824 0.0003 0.754 15.0 11.34 59.289 120 241 173 725 360 178409 525625 87800 0.0002 0.824 13.5 11.11 63.252 130 282 214 857 390 249161 734449 11234 0.0002 0.891 11.9 10.64 68.350 140 334 266 1012 420 347896 1024144 142820 0.0002 0.969 10.8 10.50 75.595 150 396 328 1196 450 485528 1430416 18072 0.0002 1.000 9.7 9.72 81.290 160 466 398 1389 475 651701 1929321 220915 0.0001 1.000 8. 8.67 93.171 165 527 459 1.000 7.9 7.93 255 Flame Front Position x Time to position x t (s) time from ignition ?t ?x ?(t 2 ) ?(t) 2 ?(t.x) Vf F(t) " . . e q )(. " . . tFq e 1/?V f MDF ? RIFT ?Full preheat Test 1 Ignition Time 364s m/s kW/m 2 (m/s) -0.5 10 20 30 40 50 60 70 80 90 100 369 5 1.000 16.9 110 382 18 1154 330 444494 1331716 127280 0.0006 1.000 15.0 15.04 41.610 120 403 39 1221 36 498429 1490841 14706 0.0004 1.000 13.5 13.49 52.387 130 436 72 1317 390 580989 1734489 171960 0.0003 1.000 11.9 11.95 61.384 140 478 114 1441 42 696309 2076481 20265 0.0002 1.000 10.8 10.83 67.520 150 527 163 1593 450 851957 2537649 240050 0.0002 1.000 9.7 9.72 74.309 160 588 224 1774 48 1057754 3147076 28516 0.0002 1.000 8. 8.67 81.318 170 659 295 1969 510 1301309 3876961 336070 0.0001 1.000 7.7 7.68 81.902 180 722 358 6. 256 Flame Front Position x Time to position x t (s) time from ignition ?t ?x ?(t 2 ) ?(t) 2 ?(t.x) Vf F(t) " . . e q )(. " . . tFq e 1/?V f MDF ? RIFT ?Full preheat Test 2 Ignition Time 357s m/s kW/m 2 (m/s) -0.5 10 20 30 40 50 60 70 80 90 100 365 8 16.9 110 382 25 1149 330 440753 1320201 126760 0.0005 1.000 15.0 15.04 43.059 120 402 45 1222 36 499372 1493284 14720 0.0003 1.000 13.5 13.49 53.630 130 438 81 1313 390 577177 1723969 171400 0.0003 1.000 11.9 11.95 59.584 140 473 116 1434 42 689102 2056356 20161 0.0002 1.000 10.8 10.83 65.530 150 523 16 1574 450 831342 2477476 237150 0.0002 1.000 9.7 9.72 72.484 160 578 221 1742 48 1018494 3034564 27990 0.0002 1.000 8. 8.67 76.870 170 641 284 1920 340 744965 3686400 201450 0.0000 1.000 7.7 7.68 173.086 180 701 34 2091 535 1463283 4372281 373715 0.0001 1.000 6. 6.70 84.507 185 749 392 6.2 Flame Front Position x Time to position t time from ignition ?t ?x ?(t 2 ) ?(t) 2 ?(t.x) V F(t) q? q?.F(t) 1/?V MDF ? RIFT ?Full preheat Test 3 Ignition Time 359s m/s kW/m 2 (m/s) -0.5 10 20 30 257 40 50 60 70 80 90 100 365 6 1.000 16.9 110 378 19 1165 330 454193 1357225 128720 0.0003 1.000 15.0 15.04 55.955 120 422 63 1257 36 529817 1580049 15163 0.0003 1.000 13.5 13.49 62.985 130 457 98 1374 390 631958 1887876 179350 0.0003 1.000 11.9 11.95 60.432 140 495 136 1476 42 728450 2178576 20731 0.0003 1.000 10.8 10.83 58.053 150 524 165 1592 450 847930 2534464 239580 0.0003 1.000 9.7 9.72 63.131 160 573 214 1734 37 602905 3006756 17028 0.0001 1.000 8. 8.67 95.726 170 637 278 1910 510 1224098 3648100 325970 0.0002 1.000 7.7 7.68 79.688 180 700 341 2085 535 1455273 4347225 37267 0.0001 1.000 6. 6.70 85.644 185 748 389 6.2 Flame Front Position x Time to position x t (s) time from ignition ?t ?x ?(t 2 ) ?(t) 2 ?(t.x) Vf F(t) " . . e q )(. " . . tFq e 1/?V f MDF ? RIFT ?Full preheat Test 4 Ignition Time 358s m/s kW/m 2 (m/s) -0.5 10 20 30 40 50 60 70 80 90 100 366 8 16.9 258 110 380 22 1148 330 439960 1317904 126640 0.0005 1.000 15.0 15.04 42.774 120 402 44 1213 36 491765 1471369 14607 0.0004 1.000 13.5 13.49 50.656 130 431 73 1304 390 569206 1700416 170210 0.0003 1.000 11.9 11.95 58.985 140 471 113 1423 42 679043 2024929 20012 0.0002 1.000 10.8 10.83 67.220 150 521 16 1573 450 830843 2474329 237050 0.0002 1.000 9.7 9.72 74.264 160 581 223 1735 48 1009691 3010225 27872 0.0002 1.000 8. 8.67 74.897 170 633 275 1903 510 1212971 3621409 324590 0.0002 1.000 7.7 7.68 73.501 180 689 331 2046 534 1399586 4186116 364846 0.0002 1.000 6. 6.70 80.027 184 724 366 1.000 6.3 Flame Front Position x Time to position x t (s) time from ignition ?t ?x ?(t 2 ) ?(t) 2 ?(t.x) Vf F(t) " . . e q )(. " . . tFq e 1/?V f MDF ? RIFT ?Full preheat Test 5 Ignition Time 367s m/s kW/m 2 (m/s) -0.5 10 20 30 40 50 60 70 80 90 100 377 10 1.000 16.9 110 395 28 1189 330 472043 1413721 131190 0.0005 1.000 15.0 15.04 44.796 120 417 50 1266 36 536030 1602756 15251 0.0003 1.000 13.5 13.49 54.896 130 454 87 1370 390 629006 1876900 178920 0.0002 1.000 11.9 11.95 64.133 140 499 132 1504 42 758718 2262016 21153 0.0002 1.000 10.8 10.83 69.702 150 551 184 1664 450 929598 2768896 250750 0.0002 1.000 9.7 9.72 75.944 160 614 247 1848 48 1147086 3415104 29700 0.0002 1.000 8. 8.67 81.268 170 683 316 2050 509 1410494 4202500 349137 0.0001 1.000 7.7 7.68 85.539 259 179 753 386 1.000 6.5 Flame Front Position x Time to position x t (s) time from ignition ?t ?x ?(t 2 ) ?(t) 2 ?(t.x) Vf F(t) " . . e q )(. " . . tFq e 1/?V f MDF ? RIFT ?Full preheat Test 6 Ignition Time 358s m/s kW/m 2 (m/s) -0.5 10 20 30 40 50 60 70 80 90 100 366 8 1.000 16.9 110 381 23 1151 330 442333 1324801 126990 0.0005 1.000 15.0 15.04 43.910 120 404 46 1219 36 496733 1485961 14681 0.0004 1.000 13.5 13.49 51.628 130 434 76 1328 390 591672 1763584 173500 0.0002 1.000 11.9 11.95 66.566 140 490 132 1454 42 709356 2114116 20452 0.0002 1.000 10.8 10.83 69.602 150 530 17 1606 450 864396 2579236 241860 0.0002 1.000 9.7 9.72 69.602 160 586 228 1788 48 1075880 3196944 28750 0.0001 1.000 8. 8.67 84.886 170 672 314 1972 510 1304776 3888784 336520 0.0002 1.000 7.7 7.68 81.560 180 714 356 2160 54 1560456 4665600 38982 0.0002 1.000 6. 6.70 71.784 190 774 41 1.000 6.0 260 Melteca/ MDF Flame Front Position x Time to position x t (s) time from ignition ?t ?x ?(t 2 ) ?(t) 2 ?(t.x) V f F(t) " . . e q )(. " . . tFq e 1/?V f LIFT Melteca /MDF- test 1 Ignition Time 498s m/s kW/m 2 (m/s) -0.5 50 75 0.000 35.4 0.00 100 0.000 35.0 0.00 125 0.000 33.7 0.00 150 0.000 32.4 0.00 175 0.000 30.7 0.00 200 0.000 29.0 0.00 225 0.000 27.4 0.00 250 0.000 25.7 0.00 275 0.000 23.3 0.00 300 518 20 0.962 20.9 20.10 325 569 71 1697 975 964185 2879809 553825 0.0005 1.000 19.2 19.20 42.980 350 610 112 1966 1050 1315230 3865156 693550 0.0002 1.000 17.5 17.50 70.183 375 787 289 2286 1125 1781790 5225796 864225 0.0002 1.000 14.9 14.90 75.594 400 889 391 2840 1200 2764586 8065600 1145425 0.0001 1.000 12.3 12.30 89.829 425 1164 666 3463 1270 4133317 11992369 1477750 0.0001 1.000 10.9 10.90 107.545 445 1410 912 1.000 9.8 9.78 Flame Front Position x Time to position x t (s) time from ignition ?t ?x ?(t 2 ) ?(t) 2 ?(t.x) V f F(t) " . . e q )(. " . . tFq e 1/?V f LIFT Melteca /MDF- test 2 Ignition Time 518s m/s kW/m 2 (m/s) -0.5 50 75 100 563 45 1.000 35.0 35.00 125 566 48 1702 375 965654 2896804 213000 0.00475 1.000 33.7 33.70 14.514 261 150 573 55 1720 450 986246 2958400 258375 0.00333 1.000 32.4 32.40 17.333 175 581 63 1746 525 1016354 3048516 306025 0.00261 1.000 30.7 30.70 19.574 200 592 74 1773 600 1048025 3143529 355075 0.00261 1.000 29.0 29.00 19.574 225 600 82 1802 675 1082564 3247204 405900 0.00277 1.000 27.4 27.35 19.013 250 610 92 1830 750 1116500 3348900 458000 0.00250 1.000 25.7 25.70 20.000 275 620 102 1865 825 1159725 3478225 513500 0.00197 1.000 23.3 23.30 22.509 300 635 117 1969 900 1297421 3876961 593050 0.00046 1.000 20.9 20.90 46.589 325 714 196 2134 975 1529246 4553956 697300 0.00033 1.000 19.2 19.20 54.798 350 785 267 2387 1050 1914565 5697769 839800 0.00028 1.000 17.5 17.50 59.323 375 888 370 2743 1125 2549669 7524049 1035750 0.00017 1.000 14.9 14.90 76.459 400 1070 552 3196 1200 3466088 10214416 1287150 0.00014 1.000 12.3 12.30 83.688 425 1238 720 3826 1265 4981868 14638276 1622070 0.00009 1.000 10.9 10.90 108.058 440 1518 1000 1.000 10.1 10.06 Flame Front Position x Time to position x t (s) time from ignition ?t ?x ?(t 2 ) ?(t) 2 ?(t.x) V f F(t) " . . e q )(. " . . tFq e 1/?V f LIFT Melteca /MDF- test 3 Ignition Time 60 m/s kW/m 2 (m/s) -0.5 50 75 0.000 35.4 100 0.000 35.0 125 0.000 33.7 150 0.000 32.4 175 0.000 30.7 200 0.000 29.0 225 0.000 27.4 250 0.000 25.7 275 580 62 1.000 23.3 300 605 87 1842 900 1134074 3392964 554525 0.0006 1.000 20.9 20.90 40.039 325 657 139 1987 975 1323299 3948169 648775 0.0004 1.000 19.2 19.20 49.135 350 725 207 2210 1050 1642858 4884100 777775 0.0003 1.000 17.5 17.50 58.888 262 375 828 310 2638 1125 2388434 6959044 998250 0.0001 1.000 14.9 14.90 87.402 400 1085 567 3383 1200 4023709 11444689 1369250 0.0001 1.000 12.3 12.30 114.062 425 1470 952 1.000 10.9 10.90 445 1675 1157 1.000 9.8 9.78 Flame Front Position x Time to position x t (s) time from ignition ?t ?x ?(t 2 ) ?(t) 2 ?(t.x) V f F(t) " . . e q )(. " . . tFq e 1/?V f RIFT Melteca/MDF low preheat-test 1 Ignition Time 75s m/s kW/m 2 (m/s) -0.5 10 75 0 0.372 34.25 12.74 20 78 3 234 60 18270 54756 4740 0.0033 0.379 33.40 12.67 17.321 30 81 6 243 90 19701 59049 7350 0.0033 0.386 31.99 12.36 17.321 40 84 9 260 120 22642 67600 10540 0.0013 0.394 29.45 11.59 27.860 50 95 20 351 150 45665 123201 18430 0.0002 0.418 27.00 11.30 72.284 60 172 97 472 180 80634 222784 29420 0.0002 0.563 24.80 13.97 76.114 70 205 130 628 210 134610 394384 44750 0.0003 0.615 22.75 13.98 63.132 80 251 176 756 240 195026 571536 61430 0.0002 0.680 20.62 14.03 68.932 90 300 225 896 270 272026 802816 81580 0.0002 0.744 18.39 13.67 68.577 100 345 270 1069 300 388801 1142761 108140 0.0002 0.797 16.28 12.99 79.721 110 424 349 1331 330 614645 1771561 148580 0.0001 0.884 14.50 12.82 105.439 120 562 487 1.000 12.91 12.91 Flame Front Position x Time to position t time from ignition ?t ?x ?(t 2 ) ?(t) 2 ?(t.x) V F(t) q? q?.F(t) 1/?V RIFT Melteca/MDF low preheat-test 2 Ignition Time 48s m/s kW/m 2 (m/s) -0.5 10 48 0 34.25 20 76 28 227 60 18689 51529 5090 0.00036 0.374 33.40 12.50 52.443 30 103 55 321 90 36549 103041 10290 0.00030 0.436 31.99 13.94 57.761 40 142 94 415 120 59673 172225 17270 0.00030 0.512 29.45 15.07 58.139 263 50 170 122 505 150 86313 255025 25760 0.00039 0.560 27.00 15.12 50.578 60 193 145 595 180 119973 354025 36320 0.00032 0.596 24.80 14.79 56.292 70 232 184 683 210 157637 466489 48460 0.00030 0.654 22.75 14.88 57.388 80 258 210 786 240 208004 617796 63520 0.00031 0.690 20.62 14.22 56.899 90 296 248 948 270 309416 898704 86680 0.00014 0.739 18.39 13.58 85.095 100 394 346 1218 300 521636 1483524 124120 0.00009 0.852 16.28 13.88 108.135 110 528 480 1522 330 794020 2316484 169480 0.00009 0.987 14.50 14.30 103.010 120 600 552 1848 359 1157184 3415104 222960 0.00010 1.000 12.91 12.91 101.790 129 720 672 1.000 12.97 12.97 Flame Front Position x Time to position x t (s) time from ignition ?t ?x ?(t 2 ) ?(t) 2 ?(t.x) V f F(t) " . . e q )(. " . . tFq e 1/?V f RIFT Melteca/MDF- low preheat-test 3 Ignition Time 60s m/s kW/m 2 (m/s) -0.5 10 60 0 20 94 34 275 60 27077 75625 6110 0.0003 0.416 33.40 13.90 55.348 30 121 61 358 90 43926 128164 11230 0.0004 0.472 31.99 15.11 49.583 40 143 83 453 120 70811 205209 18800 0.0003 0.513 29.45 15.12 59.508 50 189 129 558 150 107246 311364 28730 0.0002 0.590 27.00 15.94 64.547 60 226 166 664 180 148798 440896 40440 0.0003 0.645 24.80 16.01 55.267 70 249 189 758 210 193166 574564 53630 0.0003 0.677 22.75 15.41 53.716 80 283 223 859 240 249019 737881 69500 0.0003 0.722 20.62 14.89 62.621 90 327 267 985 270 327643 970225 89570 0.0002 0.776 18.39 14.28 67.845 100 375 315 1172 300 468454 1373584 118630 0.0001 0.831 16.28 13.54 86.067 110 470 410 1389 330 657461 1929321 154480 0.0001 0.931 14.50 13.50 92.160 120 544 484 1633 353 899997 2666689 193117 0.0001 1.000 12.91 12.91 107.124 123 619 559 1.000 13.35 13.35 264 Flame Front Position x Time to position x t (s) time from ignition ?t ?x ?(t 2 ) ?(t) 2 ?(t.x) V f F(t) " . . e q )(. " . . tFq e 1/?V f RIFT Melteca/MDF- low preheat-test 4 Ignition Time 60s m/s kW/m 2 (m/s) -0.5 10 60 0 153 30 12249 23409 2460 0.0002 0.333 34.64 11.52 69.142 20 93 33 273 60 26649 74529 6060 0.0003 0.414 33.67 13.94 54.863 30 120 60 355 90 43213 126025 11140 0.0004 0.470 33.00 15.52 49.583 40 142 82 448 120 69160 200704 18580 0.0003 0.512 30.46 15.59 58.500 50 186 126 518 150 90860 268324 26380 0.0003 0.586 27.92 16.35 54.365 60 190 130 596 180 119096 355216 36100 0.0005 0.592 25.63 15.17 45.071 70 220 160 664 210 149016 440896 47120 0.0003 0.637 23.57 15.01 56.605 80 254 194 779 240 205941 606841 63170 0.0002 0.684 21.51 14.72 65.625 90 305 245 928 270 293702 861184 84670 0.0002 0.750 19.28 14.46 75.990 100 369 309 1208 300 514342 1459264 123090 0.0001 0.825 17.05 14.06 110.419 110 534 474 1526 329 809446 2328676 169777 0.0001 0.992 15.14 15.02 117.028 119 623 563 1.000 15.14 15.14 Flame Front Position x Time to position x t (s) time from ignition ?t ?x ?(t 2 ) ?(t) 2 ?(t.x) V f F(t) " . . e q )(. " . . tFq e 1/?V f RIFT Melteca/MDF- low preheat-test 5 Ignition Time 65s m/s kW/m 2 (m/s) -0.5 10 65 0 20 105 40 297 60 31379 88209 6560 0.0003 0.440 33.67 14.81 56.454 30 127 62 406 90 57430 164836 12870 0.0003 0.484 33.00 15.97 60.008 40 174 109 491 120 82505 241081 20270 0.0003 0.566 30.46 17.25 58.346 50 190 125 599 150 121601 358801 30560 0.0003 0.592 27.92 16.53 57.269 60 235 170 705 180 169725 497025 43200 0.0002 0.658 25.63 16.87 67.082 70 280 215 846 210 243186 715716 60180 0.0002 0.718 23.57 16.93 69.327 80 331 266 1031 240 364361 1062961 83880 0.0001 0.781 21.51 16.81 84.687 265 90 420 355 1238 270 523130 1532644 112980 0.0001 0.880 19.28 16.97 88.610 100 487 422 1563 298 843905 2442969 157348 0.0001 0.947 17.05 16.15 118.971 108 656 591 1.000 17.05 17.05 Flame Front Position x Time to position x t (s) time from ignition ?t ?x ?(t 2 ) ?(t) 2 ?(t.x) V f F(t) " . . e q )(. " . . tFq e 1/?V f RIFT Melteca/MDF- low preheat-test 6 Ignition Time 60s m/s kW/m 2 (m/s) -0.5 10 60 20 70 70 217 60 16069 47089 4610 0.0007 0.359 33.67 12.09 37.152 30 87 87 258 90 22670 66564 8050 0.0006 0.400 33.00 13.22 39.431 40 101 101 317 120 34411 100489 13100 0.0005 0.431 30.46 13.14 46.667 50 129 129 408 150 58526 166464 21170 0.0003 0.488 27.92 13.62 62.813 60 178 178 515 180 91589 265225 31690 0.0002 0.573 25.63 14.68 63.452 70 208 208 635 210 136949 403225 45160 0.0003 0.619 23.57 14.60 59.820 80 249 249 770 240 203234 592900 62650 0.0002 0.677 21.51 14.58 73.034 90 313 313 943 270 305131 889249 86190 0.0002 0.760 19.28 14.65 81.253 100 381 381 1135 300 437611 1288225 114780 0.0002 0.838 17.05 14.29 80.052 110 441 441 1398 330 671418 1954404 155730 0.0001 0.902 15.14 13.65 101.147 120 576 576 1.000 13.54 266 Flame Front Position x Time to position x t (s) time from ignition ?t ?x ?(t 2 ) ?(t) 2 ?(t.x) V f F(t) " . . e q )(. " . . tFq e 1/?V f RIFT ? Melteca/MDF full preheat, test 1 Ignition Time 550 m/s kW/m 2 (m/s) -0.5 10 20 30 40 50 60 564 14 1.000 24.80 70 579 29 1768 210 1043962 3125824 124370 0.0003 1.000 22.75 22.75 57.555 80 625 75 1864 240 1161466 3474496 149930 0.0002 1.000 20.62 20.62 63.835 90 660 110 2193 270 1650689 4809249 200200 0.0001 1.000 18.39 18.39 129.699 100 908 358 1.000 16.28 Flame Front Position x Time to position x t (s) time from ignition ?t ?x ?(t 2 ) ?(t) 2 ?(t.x) V f F(t) " . . e q )(. " . . tFq e 1/?V f RIFT ? Melteca/MDF full preheat, test 2 Ignition Time 571 m/s kW/m 2 (m/s) -0.5 10 0 0 20 0 0 30 0 0 40 0 0 50 0 0 60 0 0 70 579 8 1.000 22.75 80 592 21 1776 240 1051730 3154176 142340 0.00077 1.000 20.62 20.62 36.056 90 605 34 1851 270 1144205 3426201 167210 0.00029 1.000 18.39 18.39 58.723 100 654 83 2006 300 1351750 4024036 202020 0.00014 1.000 16.28 16.28 85.599 110 747 176 2268 330 1737414 5143824 251610 0.00009 1.000 14.50 14.50 103.475 267 120 867 296 2539 360 2165323 6446521 306460 0.00011 1.000 12.91 12.91 96.229 130 925 354 2820 385 2664098 7952400 363070 0.00009 1.000 11.48 11.48 106.611 135 1028 457 1.000 11.72 Flame Front Position x Time to position x t (s) time from ignition ?t ?x ?(t 2 ) ?(t) 2 ?(t.x) V f F(t) " . . e q )(. " . . tFq e 1/?V f RIFT ? Melteca/MDF full preheat, test 3 Ignition Time 557 m/s kW/m 2 (m/s) -0.5 10 0 0 20 0 0 30 0 0 40 0 0 50 0 0 60 0 0 70 0 0 80 566 9 1.000 20.62 90 610 53 1824 270 1112360 3326976 164980 0.0002 1.000 18.39 18.39 64.088 100 648 91 1983 300 1317629 3932289 199450 0.0002 1.000 16.28 16.28 77.269 110 725 168 2255 330 1723453 5085025 250390 0.0001 1.000 14.50 14.50 110.254 120 882 325 2604 357 2297558 6780816 312209 0.0001 1.000 12.91 12.91 126.420 127 997 440 1.000 13.10 13.10 Flame Front Position x Time to position x t (s) time from ignition ?t ?x ?(t 2 ) ?(t) 2 ?(t.x) V f F(t) " . . e q )(. " . . tFq e 1/?V f RIFT ? Melteca/MDF full preheat, test 4 Ignition Time 549 m/s kW/m 2 (m/s) -0.5 10 0 0 20 0 0 30 0 0 40 0 0 50 0 0 268 60 0 0 70 555 6 1.000 23.57 80 559 10 1720 240 987742 2958400 138110 0.0003 1.000 21.51 21.51 56.163 90 606 57 1849 270 1147573 3418801 167660 0.0002 1.000 19.28 19.28 79.863 100 684 135 2138 300 1554196 4571044 216220 0.0001 1.000 17.05 17.05 112.291 110 848 299 2467 330 2061185 6086089 273880 0.0001 1.000 15.14 15.14 113.770 120 935 386 1.000 13.54 Flame Front Position x Time to position x t (s) time from ignition ?t ?x ?(t 2 ) ?(t) 2 ?(t.x) V f F(t) " . . e q )(. " . . tFq e 1/?V f RIFT ? Melteca/MDF full preheat, test 5 Ignition Time 565 m/s kW/m 2 (m/s) -0.5 10 0 0 20 0 0 30 0 0 40 0 0 50 0 0 60 0 0 70 0 0 80 571 6 1.000 21.51 90 579 14 1767 270 1041971 3122289 159490 0.0004 1.000 19.28 19.28 51.245 100 617 52 2036 300 1421530 4145296 206210 0.0001 1.000 17.05 17.05 123.432 110 840 275 1.000 15.14 269 Melteca/ particle board Flame Front Position x Time to position x t (s) time from ignition ?t ?x ?(t 2 ) ?(t) 2 ?(t.x) V f F(t) " . . e q )(. " . . tFq e 1/?V f Melteca/PB low preheat ISO ign data RIFT - Test 1 Ignition Time 60 m/s kW/m 2 (m/s) -0.5 10 60 0 0.332 34.25 11.38 20 66 6 204 60 14040 41616 4260 0.0011 0.349 33.40 11.64 30.551 30 78 18 233 90 18361 54289 7220 0.0009 0.379 31.99 12.12 33.922 40 89 29 289 120 28889 83521 12000 0.0004 0.405 29.45 11.92 48.819 50 122 62 368 150 47454 135424 19080 0.0003 0.474 27.00 12.80 58.318 60 157 97 477 180 78737 227529 29380 0.0003 0.538 24.80 13.33 61.708 70 198 138 585 210 116753 342225 41680 0.0003 0.604 22.75 13.73 60.568 80 230 170 693 240 162329 480249 56110 0.0003 0.651 20.62 13.42 57.899 90 265 205 895 270 283125 801025 82250 0.0001 0.698 18.39 12.84 97.367 100 400 340 1121 300 438161 1256641 114010 0.0001 0.858 16.28 13.97 100.472 110 456 396 1355 330 616937 1836025 150040 0.0002 0.916 14.50 13.28 70.558 120 499 439 1513 360 768301 2289169 182580 0.0002 0.958 12.91 12.37 71.707 130 558 498 1.000 11.48 11.48 Flame Front Position x Time to position x t (s) Time from ignition ?t ?x ?(t 2 ) ?(t) 2 ?(t.x) V f F(t) " . . e q )(. " . . tFq e 1/?V f Melteca/PB low preheat ISO ign data RIFT - Test 2 Ignition Time 60 m/s kW/m 2 (m/s) -0.5 10 60 0 0.332 34.25 11.38 20 69 9 204 60 13986 41616 4230 0.00132 0.356 33.40 11.90 27.568 30 75 15 231 90 17955 53361 7110 0.00107 0.372 31.99 11.88 30.551 40 87 27 272 120 25294 73984 11230 0.00055 0.400 29.45 11.78 42.516 50 110 50 341 150 40405 116281 17620 0.00035 0.450 27.00 12.15 53.716 270 60 144 84 426 180 62420 181476 26180 0.00032 0.515 24.80 12.77 55.764 70 172 112 523 210 93169 273529 37240 0.00032 0.563 22.75 12.80 56.240 80 207 147 643 240 142129 413449 52360 0.00021 0.617 20.62 12.73 68.467 90 264 204 825 270 237861 680625 75720 0.00013 0.697 18.39 12.82 86.449 100 354 294 1078 300 406612 1162084 109760 0.00010 0.807 16.28 13.14 99.105 110 460 400 1337 330 610445 1787569 148760 0.00012 0.920 14.50 13.34 92.910 120 523 463 1602 355 868290 2566404 190735 0.00009 0.981 12.91 12.66 104.910 125 619 559 13.22 Flame Front Position x Time to position x t (s) time from ignition ?t ?x ?(t 2 ) ?(t) 2 ?(t.x) V f F(t) " . . e q )(. " . . tFq e 1/?V f Melteca/PB low preheat ISO ign data RIFT - Test 3 Ignition Time 60 m/s kW/m 2 (m/s) -0.5 10 60 0 20 76 16 223 60 16945 49729 4730 0.0007 0.374 33.40 12.49 36.952 30 87 27 270 90 24794 72900 8410 0.0006 0.400 31.99 12.80 39.919 40 107 47 323 120 35659 104329 13340 0.0005 0.444 29.45 13.07 45.843 50 129 69 400 150 54986 160000 20570 0.0003 0.487 27.00 13.16 53.846 60 164 104 480 180 78506 230400 29380 0.0003 0.549 24.80 13.63 54.234 70 187 127 591 210 119465 349281 42130 0.0003 0.587 22.75 13.35 63.225 80 240 180 723 240 180185 522729 58930 0.0002 0.665 20.62 13.71 73.833 90 296 236 887 270 268417 786769 80940 0.0002 0.738 18.39 13.57 74.499 100 351 291 1116 300 430778 1245456 113330 0.0001 0.804 16.28 13.09 95.039 110 469 409 1349 330 623003 1819801 150170 0.0001 0.929 14.50 13.47 95.995 120 529 469 1642 360 914538 2696164 198790 0.0001 0.987 12.91 12.73 95.069 130 644 584 1854 381 1158338 3437316 236411 0.0001 1.000 11.48 11.48 114.829 131 681 621 11.90 271 Flame Front Position x Time to position x t (s) time from ignition ?t ?x ?(t 2 ) ?(t) 2 ?(t.x) V f F(t) " . . e q )(. " . . tFq e 1/?V f Melteca/PB low preheat ISO ign data RIFT - Test 4 Ignition Time 63 m/s kW/m 2 (m/s) -0.5 10 63 0 145 30 10693 21025 2270 0.0002 0.341 34.64 11.80 67.034 20 82 19 236 60 18974 55696 5000 0.0007 0.388 33.67 13.08 38.204 30 91 28 279 90 26241 77841 8610 0.0008 0.409 33.00 13.51 35.000 40 106 43 319 120 34401 101761 13070 0.0006 0.442 30.46 13.46 39.377 50 122 59 373 150 47145 139129 19040 0.0005 0.474 27.92 13.23 44.395 60 145 82 433 180 63465 187489 26420 0.0005 0.517 25.63 13.24 46.920 70 166 103 513 210 89385 263169 36480 0.0003 0.553 23.57 13.03 53.998 80 202 139 640 240 142344 409600 52260 0.0002 0.610 21.51 13.12 74.039 90 272 209 793 270 216549 628849 72540 0.0002 0.708 19.28 13.64 76.976 100 319 256 959 300 311169 919681 96860 0.0002 0.766 17.05 13.06 69.287 110 368 305 1127 330 430785 1270129 125180 0.0002 0.823 15.14 12.46 78.249 120 440 377 1369 360 643745 1874161 166210 0.0001 0.900 13.54 12.19 99.284 130 561 498 1640 390 916642 2689600 215190 0.0001 1.000 11.95 11.95 100.523 140 639 576 1916 415 1235698 3671056 266210 0.0001 1.000 10.78 10.78 101.617 145 716 653 10.78 Flame Front Position x Time to position x t (s) time from ignition ?t ?x ?(t 2 ) ?(t) 2 ?(t.x) V f F(t) " . . e q )(. " . . tFq e 1/?V f Melteca/PB low preheat ISO ign data RIFT - Test 5 Ignition Time 60 m/s kW/m 2 (m/s) -0.5 10 60 0 20 81 21 245 60 20977 60025 5340 0.0005 0.386 33.67 13.00 46.920 30 104 44 297 90 29921 88209 9220 0.0006 0.437 33.00 14.44 40.877 40 112 52 344 120 39744 118336 14000 0.0008 0.454 30.46 13.83 35.277 272 50 128 68 414 150 59204 171396 21320 0.0003 0.485 27.92 13.55 57.809 60 174 114 496 180 84296 246016 30420 0.0003 0.566 25.63 14.50 58.913 70 194 134 582 210 113708 338724 41140 0.0005 0.598 23.57 14.08 44.721 80 214 154 654 240 143948 427716 52840 0.0004 0.628 21.51 13.50 51.441 90 246 186 752 270 191576 565504 68460 0.0003 0.673 19.28 12.97 62.784 100 292 232 892 300 271096 795664 90280 0.0002 0.733 17.05 12.50 73.753 110 354 294 1075 330 394621 1155625 119620 0.0001 0.807 15.14 12.22 82.889 120 429 369 1365 358 648081 1863225 164916 0.0001 0.889 13.54 12.03 115.454 128 582 522 13.54 Flame Front Position x Time to position x t (s) time from ignition ?t ?x ?(t 2 ) ?(t) 2 ?(t.x) V f F(t) " . . e q )(. " . . tFq e 1/?V f Melteca/PB low preheat ISO ign data RIFT - Test 6 Ignition Time 60s m/s kW/m 2 (m/s) -0.5 10 60 20 79 79 228 60 17762 51984 4850 0.0007 0.381 33.67 12.84 38.685 30 89 89 269 90 24363 72361 8290 0.0009 0.405 33.00 13.36 33.212 40 101 101 318 120 34506 101124 13110 0.0005 0.431 30.46 13.13 45.234 50 128 128 419 150 62685 175561 21840 0.0002 0.485 27.92 13.55 68.406 60 190 190 536 180 100008 287296 33060 0.0002 0.591 25.63 15.15 68.659 70 218 218 667 210 150705 444889 47380 0.0003 0.633 23.57 14.93 59.083 80 259 259 793 240 214461 628849 64420 0.0002 0.690 21.51 14.85 70.310 90 316 316 945 270 303837 893025 86160 0.0002 0.763 19.28 14.70 74.507 100 370 370 1107 300 413997 1225449 111750 0.0002 0.825 17.05 14.07 72.467 110 421 421 1364 330 642470 1860496 152070 0.0001 0.880 15.14 13.32 104.821 120 573 573 13.54 273 Flame Front Position x Time to position x t (s) time from ignition ?t ?x ?(t 2 ) ?(t) 2 ?(t.x) V f F(t) " . . e q )(. " . . tFq e 1/?V f RIFT ? Melteca/part bd ? full preheat ? test 1 Ignition Time 550s m/s kW/m 2 (m/s) -0.5 10 20 30 40 50 60 567 17 1.000 24.80 70 57 27 1732 210 1000162 2999824 121450 0.0010 1.000 22.75 22.75 32.416 80 588 38 1839 240 1132949 3381921 148090 0.0002 1.000 20.62 20.62 76.266 90 674 124 1969 270 1299869 3876961 178400 0.0002 1.000 18.39 18.39 79.646 100 707 157 2129 300 1513629 4532641 213640 0.0003 1.000 16.28 16.28 60.946 110 748 198 2307 330 1785257 5322249 255220 0.0001 1.000 14.50 14.50 87.785 120 852 302 2508 355 2109872 6290064 298020 0.0001 1.000 12.91 12.91 103.113 125 908 358 13.22 Flame Front Position x Time to position x t (s) time from ignition ?t ?x ?(t 2 ) ?(t) 2 ?(t.x) V f F(t) " . . e q )(. " . . tFq e 1/?V f RIFT ? Melteca/part bd ? full preheat ? test 2 Ignition Time 542 m/s kW/m 2 (m/s) -0.5 10 0 0 20 30 0 0 40 50 0 0 60 70 0 0 0.000 22.75 274 80 560 18 1146 240 656996 1313316 97540 0.00003 1.000 20.62 20.62 193.417 90 586 44 1764 270 1038920 3111696 159340 0.00034 1.000 18.39 18.39 53.948 100 618 76 1862 300 1158284 3467044 186920 0.00028 1.000 16.28 16.28 60.123 110 65 116 2014 330 1359532 4056196 222740 0.00016 1.000 14.50 14.50 78.881 120 738 196 2198 360 1620812 4831204 265200 0.00014 1.000 12.91 12.91 85.027 130 802 260 2423 385 1967537 5870929 312025 0.00010 1.000 11.48 11.48 99.192 135 883 341 11.72 Flame Front Position x Time to position x t (s) Time from ignition ?t ?x ?(t 2 ) ?(t) 2 ?(t.x) V f F(t) " . . e q )(. " . . tFq e 1/?V f RIFT ? Melteca/part bd ? full preheat ? test 3 Ignition Time 545 m/s kW/m 2 (m/s) -0.5 10 20 30 40 50 60 70 80 0.000 20.62 90 572 27 1182 270 699284 1397124 112480 0.0000 1.000 18.39 18.39 195.681 100 610 65 1947 300 1284509 3790809 196630 0.0001 1.000 16.28 16.28 104.077 110 765 220 2239 330 1703821 5013121 248830 0.0001 1.000 14.50 14.50 113.604 120 864 319 2577 360 2230425 6640929 311070 0.0001 1.000 12.91 12.91 95.763 130 948 403 2841 389 2704041 8071281 369951 0.0001 1.000 11.48 11.48 93.179 139 1029 484 1.000 11.53 275 Flame Front Position x Time to position x t (s) Time from ignition ?t ?x ?(t 2 ) ?(t) 2 ?(t.x) V f F(t) " . . e q )(. " . . tFq e 1/?V f RIFT ? Melteca/part bd ? full preheat ? test 4 Ignition Time 545 m/s kW/m 2 (m/s) -0.5 10 20 30 40 50 60 70 80 559 14 1.000 21.51 90 583 38 1743 270 1013571 3038049 157290 0.0005 1.000 19.28 19.28 45.981 100 601 56 1863 300 1162131 3470769 187260 0.0002 1.000 17.05 17.05 73.655 110 679 134 2084 330 1468658 4343056 231270 0.0001 1.000 15.14 15.14 101.643 120 804 259 2371 360 1896001 5621641 286610 0.0001 1.000 13.54 13.54 102.879 130 888 343 2694 383 2438964 7257636 345186 0.0001 1.000 11.95 11.95 125.604 133 1002 457 1.000 11.95 Flame Front Position x Time to position x t (s) Time from ignition ?t ?x ?(t 2 ) ?(t) 2 ?(t.x) V f F(t) " . . e q )(. " . . tFq e 1/?V f RIFT ? Melteca/part bd ? full preheat ? test 5 Ignition Time 613 m/s kW/m 2 (m/s) -0.5 10 20 30 40 276 50 60 70 80 625 12 1.000 21.51 90 639 26 1921 270 1230595 3690241 173210 0.0006 1.000 19.28 19.28 40.104 100 657 44 1991 300 1322995 3964081 199660 0.0003 1.000 17.05 17.05 54.028 110 695 82 2088 330 1456370 4359744 230470 0.0003 1.000 15.14 15.14 62.864 120 736 123 2213 360 1636245 4897369 266430 0.0002 1.000 13.54 13.54 65.991 130 782 169 2390 385 1913604 5712100 307700 0.0001 1.000 11.95 11.95 98.655 135 87 259 1.000 11.95 Flame Front Position x Time to position x t (s) Time from ignition ?t ?x ?(t 2 ) ?(t) 2 ?(t.x) V f F(t) " . . e q )(. " . . tFq e 1/?V f RIFT ? Melteca/part bd ? full preheat ? test 6 Ignition Time 543 m/s kW/m 2 (m/s) -0.5 10 20 30 40 50 60 70 551 8 1.000 23.57 80 576 33 1728 240 996578 2985984 138740 0.0004 1.000 21.51 21.51 50.000 90 601 58 1821 270 1107713 3316041 164570 0.0003 1.000 19.28 19.28 58.987 100 644 101 1912 300 1220826 3655744 191860 0.0003 1.000 17.05 17.05 58.318 110 667 124 2002 330 1337106 4008004 220690 0.0004 1.000 15.14 15.14 48.480 120 691 148 2061 360 1416579 4247721 247680 0.0005 1.000 13.54 13.54 43.205 130 703 160 2141 390 1529699 4583881 278890 0.0003 1.000 11.95 11.95 55.720 140 747 204 2285 420 1749443 5221225 321220 0.0001 1.000 10.78 10.78 82.731 150 835 292 2507 447 2110859 6285049 375055 0.0001 1.000 9.61 9.61 102.362 277 157 925 382 9.02 Hardboard Flame Front Position x Time to position x t (s) time from ignition ?t ?x ?(t 2 ) ?(t) 2 ?(t.x) V f V f F(t) " . . e q )(. " . . tFq e 1/?V f LIFT ? Hardboard ? test 1 Ignition Time 606 m/s mm/s kW/m 2 (m/s) -0.5 50 0 0 75 0 0 22.05 100 0 0 21.94 125 0 0 21.10 150 0 0 20.27 175 0 0 19.66 200 0 0 19.05 225 0 0 18.25 250 612 6 1.000 17.45 17.45 275 626 20 1878 825 1176020 3526884 517150 0.0018 1.7857 1.000 15.86 15.86 23.664 300 640 34 1933 900 1246365 3736489 580925 0.0012 1.1800 1.000 14.28 14.28 29.112 325 667 61 2021 975 1364285 4084441 658675 0.0007 0.6596 1.000 12.89 12.89 38.936 350 714 108 2155 1050 1553761 4644025 756925 0.0005 0.4650 1.000 11.50 11.50 46.374 375 774 168 2325 1125 1809441 5405625 874950 0.0004 0.4064 1.000 10.17 10.17 49.603 400 837 231 2542 1200 2166406 6461764 1020725 0.0003 0.3144 1.000 8.83 8.83 56.399 425 931 325 2834 1275 2703686 8031556 1210175 0.0002 0.2160 1.000 7.59 7.59 68.036 450 1066 460 3223 1350 3506193 10387729 1457725 0.0002 0.1691 1.000 6.35 6.35 76.903 475 1226 620 3726 1425 4695788 13883076 1779050 0.0001 0.1351 1.000 5.42 5.42 86.033 500 1434 828 4359 1500 6446033 19000881 2191325 0.0001 0.1052 1.000 4.50 4.50 97.498 525 1699 1093 1.000 3.79 3.79 Flame Front Position x Time to position x t (s) time from ignition ?t ?x ?(t 2 ) ?(t) 2 ?(t.x) V f V f F(t) " . . e q )(. " . . tFq e 1/?V f LIFT ? Hardboard Ignition 608 m/s kW/m 2 (m/s) -0.5 278 ? test 2 Time 50 0 0 75 0 0 22.05 100 0 0 21.94 125 0 0 21.10 150 0 0 20.27 175 0 0 19.66 200 0 0 19.05 225 0 0 18.25 250 0 0 17.45 275 617 11 1.000 15.86 15.86 300 626 20 1887 900 1187301 3560769 566775 0.0018 1.7857 1.000 14.28 14.28 23.664 325 644 38 1936 975 1250168 3748096 630200 0.0012 1.2458 1.000 12.89 12.89 28.331 350 666 60 2010 1050 1348292 4040100 704900 0.0009 0.8794 1.000 11.50 11.50 33.722 375 700 94 2117 1125 1497557 4481689 796000 0.0006 0.5805 1.000 10.17 10.17 41.505 400 751 145 2265 1200 1716597 5130225 908850 0.0004 0.4370 1.000 8.83 8.83 47.837 425 814 208 2472 1275 2049246 6110784 1054500 0.0003 0.3166 1.000 7.59 7.59 56.200 450 907 301 1721 925 1485245 2961841 754100 0.0004 0.4487 1.000 6.35 6.35 47.206 475 1015 409 1922 1000 1852874 3694084 890275 0.0004 0.4016 1.000 5.42 5.42 49.899 500 1111 505 2126 1075 2264546 4519876 1037625 0.0004 0.3639 1.000 4.50 4.50 52.421 525 1369 763 2480 1150 3108482 6150400 1274225 0.0003 0.3057 1.000 3.79 3.79 57.192 550 1597 991 2966 1225 4424570 8797156 1597075 0.0003 0.2587 1.000 3.08 3.08 62.179 558 1665 1059 2.96 Flame Front Position x Time to position x t (s) time from ignition ?t ?x ?(t 2 ) ?(t) 2 ?(t.x) V f V f F(t) " . . e q )(. " . . tFq e 1/?V f LIFT ? Hardboard ? test 3 Ignition Time 610 m/s mm/s kW/m 2 50 75 100 279 125 150 175 200 225 250 275 617 11 1.000 15.86 15.86 300 622 16 1877 900 1174617 3523129 563625 0.0022 2.1814 1.000 14.28 14.28 21.411 325 638 32 1923 975 1233497 3697929 626000 0.0012 1.2002 1.000 12.89 12.89 28.865 350 663 57 1997 1050 1331029 3988009 700400 0.0009 0.8566 1.000 11.50 11.50 34.167 375 696 90 2099 1125 1471585 4405801 789050 0.0006 0.6450 1.000 10.17 10.17 39.376 400 740 134 2238 1200 1675220 5008644 897850 0.0005 0.4672 1.000 8.83 8.83 46.264 425 802 196 2422 1275 1965204 5866084 1032850 0.0004 0.3556 1.000 7.59 7.59 53.030 450 880 274 2666 1350 2385860 7107556 1204250 0.0003 0.2729 1.000 6.35 6.35 60.537 475 984 378 3012 1425 3060560 9072144 1437400 0.0002 0.1835 1.000 5.42 5.42 73.821 500 1148 542 3514 1500 4196084 12348196 1766950 0.0001 0.1243 1.000 4.50 4.50 89.678 525 1382 776 4082 1565 5636532 16662724 2137630 0.0001 0.0995 1.000 3.79 3.79 100.259 540 1552 946 1.000 3.36 Flame Front Position x Time to position x t (s) time from ignition ?t ?x ?(t 2 ) ?(t) 2 ?(t.x) V f F(t) " . . e q )(. " . . tFq e 1/?V f RIFT ? Hardboard ? low preheat ? ISO ign data ? test 1 Ignition Time 80 m/s kW/m 2 (m/s) -0.5 10 80 0 0.291 33.88 20 83 3 250 60 20858 62500 5070 0.0028 0.296 32.87 9.75 18.772 30 7 7 259 90 22379 67081 783 0.0032 0.304 32.16 9.76 17.638 40 89 9 272 120 24706 73984 10970 0.0020 0.307 29.67 9.11 22.278 50 96 16 29 150 2858 85264 14780 0.0011 0.319 27.1 8.66 30.246 60 107 27 321 180 34589 103041 19480 0.0009 0.337 24.95 8.40 33.166 70 118 38 359 210 43329 128881 25400 0.0007 0.354 22.99 8.13 36.952 80 134 54 405 240 55289 164025 32750 0.0006 0.377 21.03 7.92 41.884 280 90 153 73 459 270 70949 210681 41690 0.0005 0.403 18.88 7.60 43.589 100 172 92 525 300 92993 275625 52970 0.0004 0.427 16.73 7.14 48.772 110 200 120 602 330 122484 362404 66800 0.0003 0.460 15.06 6.93 53.862 120 230 150 699 360 165261 488601 84570 0.0003 0.494 13.89 6.85 58.903 130 269 189 803 390 217677 644809 105130 0.0003 0.534 12.71 6.78 60.857 140 304 224 922 420 286578 850084 129880 0.0002 0.567 11.59 6.58 63.410 150 349 269 1054 450 375018 1110916 159070 0.0002 0.608 10.48 6.37 69.702 160 401 321 1198 480 483306 1435204 192670 0.0002 0.652 9.39 6.12 70.386 170 448 368 1360 510 622626 1849600 232300 0.0002 0.689 8.33 5.74 74.423 180 511 431 1531 540 789009 2343961 276820 0.0002 0.736 7.27 5.35 78.743 190 572 492 1720 570 994074 2958400 328060 0.0002 0.778 6.55 5.10 79.386 200 637 557 1915 600 1231389 3667225 384340 0.0001 0.821 5.83 4.79 81.866 210 706 626 2118 630 1504830 4485924 446160 0.0001 0.865 5.17 4.47 83.066 220 775 695 2329 660 1818165 5424241 513800 0.0001 0.906 4.58 4.15 84.273 230 848 768 2557 690 2192085 6538249 589700 0.0001 0.948 3.99 3.78 89.262 240 934 854 2809 720 2646189 7890481 675950 0.0001 0.995 3.77 3.75 94.629 250 1027 947 3079 750 3177009 9480241 771590 0.0001 1.000 3.54 3.54 95.919 260 1118 1038 3341 780 3735069 11162281 870350 0.0001 1.000 3.29 3.29 92.014 270 1196 1116 3614 805 4370340 13060996 971100 0.0001 1.000 3.29 3.29 111.413 275 1300 1220 1.000 3.09 3.09 Flame Front Position x Time to position x t (s) time from ignition ?t ?x ?(t 2 ) ?(t) 2 ?(t.x) V f F(t) " . . e q )(. " . . tFq e 1/?V f RIFT ? Hardboard ? low preheat ? ISO ign data ? test 2 Ignition Time 80 m/s kW/m 2 (m/s) -0.5 10 80 0 0.291 33.88 20 82 2 246 60 20180 60516 4960 0.0050 0.295 32.87 9.69 14.142 30 4 4 252 90 21176 63504 760 0.0050 0.298 32.16 9.59 14.142 40 86 6 26 120 2291 68644 10560 0.0023 0.302 29.67 8.95 20.817 50 92 12 279 150 26061 77841 14100 0.0013 0.312 27.1 8.48 27.568 60 101 21 305 180 31209 93025 18500 0.0010 0.327 24.95 8.16 31.675 281 70 112 32 342 210 39386 116964 24220 0.0007 0.344 22.99 7.92 37.702 80 129 49 386 240 50210 148996 31210 0.0006 0.370 21.03 7.78 40.626 90 145 65 442 270 65890 195364 40170 0.0005 0.392 18.88 7.40 44.395 100 168 88 509 300 87665 259081 51410 0.0004 0.422 16.73 7.06 50.578 110 196 116 592 330 118624 350464 65720 0.0003 0.456 15.06 6.86 54.813 120 228 148 687 360 159569 471969 83110 0.0003 0.491 13.89 6.82 57.899 130 263 183 798 390 215402 636804 104530 0.0003 0.528 12.71 6.71 62.985 140 307 227 927 420 290867 859329 130720 0.0002 0.570 11.59 6.61 68.603 150 357 277 1086 450 399782 1179396 164050 0.0002 0.615 10.48 6.44 76.043 160 422 342 1263 480 539789 1595169 203350 0.0002 0.669 9.39 6.28 79.694 170 484 404 986 340 418740 972196 150600 0.0004 0.716 8.33 5.97 49.363 180 555 475 1121 370 549005 1256641 183820 0.0004 0.767 7.27 5.58 53.441 190 635 555 1274 400 718306 1623076 223070 0.0003 0.820 6.55 5.37 57.725 200 718 638 1440 430 926318 2073600 267730 0.0003 0.872 5.83 5.08 61.917 210 812 732 1619 460 1182789 2621161 318570 0.0002 0.927 5.17 4.80 66.295 226 911 831 0.982 4.23 4.15 Flame Front Position x Time to position x t (s) time from ignition ?t ?x ?(t 2 ) ?(t) 2 ?(t.x) V f F(t) " . . e q )(. " . . tFq e 1/?V f RIFT ? Hardboard ? low preheat ? ISO ign data ? test 3 Ignition Time 69 m/s kW/m 2 (m/s) -0.5 10 80 0 0.291 33.88 20 82 2 246 60 20180 60516 4960 0.0050 0.295 32.87 9.69 14.142 30 4 4 253 90 21349 64009 764 0.0039 0.298 32.16 9.59 15.916 40 87 7 260 120 22546 67600 10450 0.0039 0.304 29.67 9.01 15.916 50 89 9 275 150 25291 75625 13870 0.0015 0.307 27.17 8.34 26.247 60 99 19 299 180 30043 89401 18160 0.0009 0.324 24.95 8.08 33.212 70 111 31 339 210 38763 114921 24030 0.0007 0.343 22.99 7.88 38.987 80 129 49 390 240 51462 152100 31590 0.0005 0.370 21.03 7.78 44.202 90 150 70 459 270 71541 210681 41820 0.0004 0.399 18.88 7.53 50.759 282 100 180 100 533 300 96109 284089 53830 0.0004 0.437 16.73 7.30 51.628 110 203 123 617 330 128365 380689 68410 0.0004 0.464 15.06 6.98 52.151 120 234 154 707 360 168865 499849 85510 0.0003 0.498 13.89 6.91 57.933 130 270 190 818 390 226252 669124 107140 0.0002 0.535 12.71 6.80 63.351 140 314 234 932 420 292600 868624 131260 0.0003 0.577 11.59 6.69 62.621 150 348 268 1080 450 394424 1166400 163040 0.0002 0.607 10.48 6.36 73.537 160 418 338 869 370 306437 755161 125260 0.0003 0.665 9.39 6.25 55.007 170 481 401 1017 400 420009 1034289 156910 0.0003 0.714 8.33 5.95 59.422 180 554 474 1170 430 556502 1368900 192290 0.0002 0.766 7.27 5.57 63.835 190 625 545 1335 460 721877 1782225 232510 0.0002 0.814 6.55 5.33 67.790 200 718 638 1522 490 938190 2316484 280250 0.0002 0.872 5.83 5.08 72.420 208 828 748 0.936 5.30 4.97 Flame Front Position x Time to position x t (s) time from ignition ?t ?x ?(t 2 ) ?(t) 2 ?(t.x) V f F(t) " . . e q )(. " . . tFq e 1/?V f RIFT ? Hardboard ? low preheat ? ISO ign data ? test 4 Ignition Time 66 m/s kW/m 2 (m/s) -0.5 10 80 0 0.291 33.88 20 83 3 248 60 20514 61504 5010 0.0039 0.296 32.87 9.75 15.916 30 85 5 255 90 21683 65025 7690 0.0050 0.300 32.16 9.65 14.142 40 7 7 26 120 2344 70225 10680 0.0023 0.304 29.67 9.01 20.817 50 93 13 283 150 26827 80089 14310 0.0012 0.314 27.1 8.53 28.577 60 103 23 314 180 33182 98596 19090 0.0008 0.330 24.95 8.24 35.590 70 118 38 356 210 42758 126736 25240 0.0006 0.354 22.99 8.13 40.026 80 135 55 409 240 56485 167281 33100 0.0005 0.378 21.03 7.95 43.669 90 156 76 470 270 74602 220900 42740 0.0005 0.406 18.88 7.67 46.920 100 179 99 544 300 100058 295936 54930 0.0004 0.435 16.73 7.28 51.628 110 209 129 626 330 132366 391876 69450 0.0003 0.470 15.06 7.09 54.317 120 238 158 719 360 174309 516961 86910 0.0003 0.502 13.89 6.97 56.184 283 130 272 192 825 390 229853 680625 108020 0.0003 0.537 12.71 6.82 62.189 140 315 235 943 420 299945 889249 132860 0.0002 0.578 11.59 6.70 64.814 150 356 276 1083 450 395705 1172889 163420 0.0002 0.614 10.48 6.43 69.919 160 412 332 1247 480 525921 1555009 200750 0.0002 0.661 9.39 6.20 78.526 170 479 399 1439 510 699489 2070721 245990 0.0001 0.712 8.33 5.93 82.465 180 548 468 1652 540 920370 2729104 298820 0.0001 0.762 7.27 5.54 85.483 190 625 545 1889 570 1203585 3568321 360590 0.0001 0.814 6.55 5.33 91.758 200 716 636 2154 600 1564250 4639716 432680 0.0001 0.871 5.83 5.08 96.970 210 813 733 2460 628 2040386 6051600 516888 0.0001 0.928 5.17 4.80 109.663 218 931 851 0.993 4.70 4.67 Flame Front Position x Time to position x t (s) time from ignition ?t ?x ?(t 2 ) ?(t) 2 ?(t.x) V f F(t) " . . e q )(. " . . tFq e 1/?V f RIFT ? Hardboard ? low preheat ? ISO ign data ? test 5 Ignition Time 67 m/s kW/m 2 (m/s) -0.5 10 80 0 0.291 33.88 20 83 3 250 60 20858 62500 5070 0.0028 0.296 32.87 9.75 18.772 30 7 7 262 90 22922 68644 795 0.0022 0.304 32.16 9.76 21.257 40 92 12 279 120 26033 77841 11290 0.0015 0.312 29.67 9.26 25.720 50 100 20 303 150 30785 91809 15340 0.0010 0.325 27.17 8.84 30.950 60 111 31 336 180 37946 112896 20410 0.0008 0.343 24.95 8.55 35.440 70 125 45 377 210 47827 142129 26690 0.0007 0.364 22.99 8.37 38.759 80 141 61 427 240 61427 182329 34520 0.0006 0.386 21.03 8.13 42.514 90 161 81 488 270 80398 238144 44370 0.0004 0.413 18.88 7.80 47.532 100 186 106 563 300 107173 316969 56850 0.0004 0.444 16.73 7.42 52.513 110 216 136 648 330 141768 419904 71880 0.0003 0.478 15.06 7.20 54.772 120 246 166 744 360 186696 553536 89940 0.0003 0.510 13.89 7.09 57.525 130 282 202 847 390 241801 717409 110840 0.0003 0.547 12.71 6.95 60.417 140 319 239 972 420 318926 944784 136970 0.0002 0.581 11.59 6.74 67.023 150 371 291 1110 450 415802 1232100 167510 0.0002 0.627 10.48 6.57 71.074 160 420 340 1265 480 538717 1600225 203430 0.0002 0.667 9.39 6.26 71.792 284 170 474 394 1424 510 681976 2027776 243180 0.0002 0.709 8.33 5.90 74.166 180 530 450 1598 540 858412 2553604 288840 0.0002 0.749 7.27 5.45 77.517 190 594 514 1789 570 1075961 3200521 341260 0.0001 0.793 6.55 5.20 82.195 200 665 585 2001 600 1345625 4004001 401680 0.0001 0.839 5.83 4.89 86.047 210 742 662 2228 630 1666830 4963984 469440 0.0001 0.886 5.17 4.59 88.320 220 821 741 2451 660 2013149 6007401 540680 0.0001 0.932 4.58 4.27 85.536 230 888 808 2682 690 2409314 7193124 618380 0.0001 0.970 3.99 3.87 87.381 240 973 893 2936 720 2890898 8620096 706510 0.0001 1.000 3.77 3.77 96.828 250 1075 995 3206 750 3443318 10278436 803350 0.0001 1.000 3.54 3.54 96.346 260 1158 1078 3479 780 4049105 12103441 906250 0.0001 1.000 3.29 3.29 92.479 270 1246 1166 1.000 3.29 3.29 Flame Front Position x Time to position x t (s) time from ignition ?t ?x ?(t 2 ) ?(t) 2 ?(t.x) V f F(t) " . . e q )(. " . . tFq e 1/?V f RIFT ? hardboard ? Full preheat ? test 1 Ignition Time 985 m/s kW/m 2 (m/s) -0.5 10 20 30 40 50 60 70 80 90 100 11 120 986 1 1.000 13.89 130 989 4 2975 390 2950317 8850625 386890 0.0013 1.000 12.71 12.71 27.860 140 1000 15 3007 420 3014445 9042049 421270 0.0007 1.000 11.59 11.59 38.447 285 150 1018 33 3061 450 3124173 9369721 459580 0.0005 1.000 10.48 10.48 46.572 160 1043 58 3136 480 3279798 9834496 502330 0.0003 1.000 9.39 9.39 53.519 170 1075 90 3227 510 3473355 10413529 549250 0.0003 1.000 8.33 8.33 57.454 180 1109 124 3339 540 3719531 11148921 601820 0.0002 1.000 7.27 7.27 63.482 190 1155 170 3459 570 3991931 11964681 658070 0.0002 1.000 6.55 6.55 65.628 200 1195 210 3597 600 4317059 12938409 720320 0.0002 1.000 5.83 5.83 68.015 210 1247 262 3744 630 4678238 14017536 787310 0.0002 1.000 5.17 5.17 73.153 220 1302 317 3924 660 5140838 15397776 864560 0.0002 1.000 4.58 4.58 80.263 230 1375 390 4127 690 5688329 17032129 950690 0.0001 1.000 3.99 3.99 86.026 240 1450 465 4343 720 6297449 18861649 1043750 0.0001 1.000 3.77 3.77 84.591 250 1518 533 4572 750 6979640 20903184 1144540 0.0001 1.000 3.54 3.54 87.949 260 1604 619 4813 780 7736621 23164969 1253110 0.0001 1.000 3.29 3.29 93.006 270 1691 706 3.29 Flame Front Position x Time to position x t (s) time from ignition ?t ?x ?(t 2 ) ?(t) 2 ?(t.x) V f F(t) " . . e q )(. " . . tFq e 1/?V f RIFT ? hardboard ? Full preheat ? test 2 Ignition Time 987 m/s kW/m 2 (m/s) -0.5 10 20 30 40 50 60 70 80 90 100 11 120 130 993 6 12.71 0.00 #DIV/0! 140 1004 17 3017 420 3034465 9102289 422650 0.0007 1.000 11.59 11.59 36.952 286 150 1020 33 3063 450 3127937 9381969 459800 0.0006 1.000 10.48 10.48 41.884 160 1039 52 3127 480 3260545 9778129 500800 0.0004 1.000 9.39 9.39 49.343 170 1068 81 2107 340 2220145 4439449 347800 0.0001 1.000 8.33 8.33 82.411 180 1101 114 2169 370 2352825 4704561 379740 0.0001 1.000 7.27 7.27 83.614 190 1140 153 2241 400 2511801 5022081 414780 0.0001 1.000 6.55 6.55 84.991 200 1186 199 2326 430 2706196 5410276 453800 0.0001 1.000 5.83 5.83 86.589 210 1233 246 2419 460 2926885 5851561 496130 0.0001 1.000 5.17 5.17 88.303 220 1286 299 2519 490 3174085 6345361 541850 0.0001 1.000 4.58 4.58 90.111 230 1350 363 2636 520 3476296 6948496 593420 0.0001 1.000 3.99 3.99 92.186 240 1408 421 2758 550 3804964 7606564 648420 0.0001 1.000 3.77 3.77 94.289 250 1477 490 2885 580 4163993 8323225 707170 0.0001 1.000 3.54 3.54 96.441 260 1556 569 3033 610 4602665 9199089 773810 0.0001 1.000 3.29 3.29 98.890 270 1626 639 3182 640 5065012 10125124 843580 0.0001 1.000 3.29 3.29 101.280 280 1674 687 2.89 Flame Front Position x Time to position x t (s) time from ignition ?t ?x ?(t 2 ) ?(t) 2 ?(t.x) V f F(t) " . . e q )(. " . . tFq e 1/?V f RIFT ? hardboard ? Full preheat ? test 3 Ignition Time 987 m/s kW/m 2 (m/s) -0.5 10 20 30 40 50 60 70 80 90 100 11 120 130 994 7 12.71 287 140 1001 14 3012 420 3024326 9072144 421910 0.0008 1.000 11.59 11.59 34.766 150 1017 30 3056 450 3113734 9339136 458770 0.0005 1.000 10.48 10.48 43.142 160 1038 51 2055 370 2111733 4223025 318630 0.0001 1.000 9.39 9.39 103.931 170 1066 79 2104 400 2213800 4426816 347300 0.0001 1.000 8.33 8.33 105.149 180 1100 113 2166 430 2346356 4691556 379220 0.0001 1.000 7.27 7.27 106.678 190 1077 90 2177 460 2369929 4739329 402630 0.0001 1.000 6.55 6.55 107.149 200 1182 195 2259 490 2557053 5103081 441030 0.0001 1.000 5.83 5.83 108.992 210 1227 240 2409 520 2902653 5803281 494070 0.0001 1.000 5.17 5.17 112.494 220 1290 303 3503 550 4141825 12271009 659790 0.0003 1.000 4.58 4.58 54.129 230 1358 371 3636 580 4484408 13220496 724580 0.0003 1.000 3.99 3.99 59.901 240 1423 436 3787 610 4881129 14341369 794700 0.0002 1.000 3.77 3.77 63.872 250 1505 518 3950 640 5334438 15602500 871070 0.0002 1.000 3.54 3.54 68.585 260 1585 598 4135 670 5869275 17098225 955550 0.0002 1.000 3.29 3.29 72.783 270 1691 704 3.29 Flame Front Position x Time to position x t (s) time from ignition ?t ?x ?(t 2 ) ?(t) 2 ?(t.x) V f F(t) " . . e q )(. " . . tFq e 1/?V f RIFT ? hardboard ? Full preheat ? test 4 Ignition Time 982 m/s kW/m 2 (m/s) -0.5 10 20 30 40 50 60 70 80 90 100 11 288 120 986 4 13.89 130 988 6 2980 390 2960376 8880400 387600 0.0008 1.000 12.71 12.71 34.833 140 1006 24 3016 420 3032664 9096256 422580 0.0006 1.000 11.59 11.59 41.255 150 1022 40 3073 450 3148545 9443329 461340 0.0005 1.000 10.48 10.48 44.395 160 1045 63 3137 480 3281409 9840769 502400 0.0004 1.000 9.39 9.39 49.004 170 1070 88 3219 510 3455741 10361961 547820 0.0003 1.000 8.33 8.33 54.524 180 1104 122 3316 540 3667880 10995856 597600 0.0003 1.000 7.27 7.27 60.031 190 1142 160 3410 570 3877876 11628100 648500 0.0003 1.000 6.55 6.55 55.418 200 1164 182 3532 600 4162136 12475024 707240 0.0002 1.000 5.83 5.83 67.212 210 1226 244 3668 630 4491256 13454224 771420 0.0002 1.000 5.17 5.17 75.595 220 1278 296 3835 660 4907921 14707225 844750 0.0002 1.000 4.58 4.58 72.458 230 1331 349 4000 690 5339726 16000000 921130 0.0002 1.000 3.99 3.99 75.215 240 1391 409 4186 720 5849738 17522596 1005970 0.0001 1.000 3.77 3.77 81.677 250 1464 482 4388 750 6428266 19254544 1098420 0.0001 1.000 3.54 3.54 84.273 260 1533 551 4600 780 7062994 21160000 1197390 0.0001 1.000 3.29 3.29 83.367 270 1603 621 4829 806 7785947 23319241 1298658 0.0001 1.000 3.29 3.29 100.786 276 1693 711 3.05 Flame Front Position x Time to position x t (s) time from ignition ?t ?x ?(t 2 ) ?(t) 2 ?(t.x) V f F(t) " . . e q )(. " . . tFq e 1/?V f RIFT ? hardboard ? Full preheat ? test 5 Ignition Time 1054 m/s kW/m 2 (m/s) -0.5 10 20 30 40 50 60 70 80 90 289 100 11 120 130 1062 8 12.71 140 1069 15 3215 420 3445661 10336225 450320 0.0009 1.000 11.59 11.59 33.889 150 1084 30 3257 450 3536633 10608049 488900 0.0006 1.000 10.48 10.48 41.975 160 1104 50 3316 480 3666256 10995856 531000 0.0005 1.000 9.39 9.39 46.969 170 1128 74 3392 510 3836800 11505664 577200 0.0004 1.000 8.33 8.33 53.095 180 1160 106 3488 540 4057984 12166144 628560 0.0003 1.000 7.27 7.27 60.123 190 1200 146 3598 570 4318244 12945604 684400 0.0003 1.000 6.55 6.55 62.457 200 1238 184 3728 600 4636744 13897984 746500 0.0002 1.000 5.83 5.83 67.352 210 1290 236 3888 630 5046344 15116544 817700 0.0002 1.000 5.17 5.17 78.385 220 1360 306 4074 660 5541476 16597476 897620 0.0001 1.000 4.58 4.58 81.881 230 1424 370 4284 690 6127376 18352656 986720 0.0001 1.000 3.99 3.99 83.768 240 1500 446 4522 715 6831380 20448484 1079030 0.0001 1.000 3.77 3.77 108.757 245 1598 544 3.66 290 Particle board Flame Front Position x Time to position x t (s) time from ignition ?t ?x ?(t 2 ) ?(t) 2 ?(t.x) V f F(t) " . . e q )(. " . . tFq e 1/?V f LIFT-Pynefloor particle Board ? test 1 Ignition Time 396 m/s kW/m 2 (m/s) -0.5 50 75 100 125 404 8 1.000 24.24 150 409 13 1230 450 504386 1512900 184825 0.0038 1.000 23.10 23.10 16.267 175 417 21 1254 525 524354 1572516 219925 0.0026 1.000 21.67 21.67 19.574 200 428 32 1282 600 548042 1643524 256900 0.0025 1.000 20.25 20.25 20.033 225 437 41 1311 675 573069 1718721 295425 0.0028 1.000 19.18 19.18 18.974 250 446 50 1358 750 615510 1844164 340450 0.0012 1.000 18.11 18.11 28.813 275 475 79 1432 825 685662 2050624 395425 0.0008 1.000 16.40 16.40 36.125 300 511 115 1533 900 785955 2350089 461700 0.0007 1.000 14.70 14.70 37.947 325 547 151 1650 975 910794 2722500 538275 0.0006 1.000 12.98 12.98 40.332 350 592 196 1804 1050 1091898 3254416 634350 0.0004 1.000 11.27 11.27 49.034 375 665 269 2019 1125 1373333 4076361 761375 0.0003 1.000 9.94 9.94 58.503 400 762 366 2311 1200 1804325 5340721 929875 0.0002 1.000 8.62 8.62 66.325 425 884 488 2658 1275 2386244 7064964 1135900 0.0002 1.000 7.44 7.44 70.717 450 1012 616 3039 1335 3112049 9235521 1356880 0.0001 1.000 6.25 6.25 86.096 460 1143 747 1.000 5.89 Flame Front Position x Time to position x t (s) time from ignition ?t ?x ?(t 2 ) ?(t) 2 ?(t.x) V f F(t) " . . e q )(. " . . tFq e 1/?V f LIFT-Pynefloor particle Board ? test 2 Ignition Time 391 m/s kW/m 2 (m/s) -0.5 50 391 291 75 391 100 391 125 391 150 391 175 391 200 391 225 395 4 1.000 19.18 250 400 9 1223 750 499209 1495729 306575 0.0013 1.000 18.11 18.11 27.692 275 428 37 1281 825 548393 1640961 353600 0.0009 1.000 16.40 16.40 32.575 300 453 62 1386 900 643418 1920996 417725 0.0006 1.000 14.70 14.70 40.039 325 505 114 1546 975 805978 2390116 505825 0.0004 1.000 12.98 12.98 52.416 350 588 197 1743 1050 1023269 3038049 613675 0.0003 1.000 11.27 11.27 54.040 375 650 259 1996 1125 1342808 3984016 752750 0.0003 1.000 9.94 9.94 59.017 400 758 367 2280 1200 1757448 5198400 917550 0.0002 1.000 8.62 8.62 66.641 425 872 481 2622 1256 1334948 6874884 673800 0.0004 1.000 7.44 7.44 47.504 431 992 601 1.000 7.15 Flame Front Position x Time to position x t (s) time from ignition ?t ?x ?(t 2 ) ?(t) 2 ?(t.x) V f F(t) " . . e q )(. " . . tFq e 1/?V f LIFT-Pynefloor particle Board ? test 3 Ignition Time 393 m/s kW/m 2 (m/s) -0.5 50 75 100 125 150 175 200 225 400 7 19.18 250 407 14 1226 750 501210 1503076 306975 0.0026 1.000 18.11 18.11 19.717 275 419 26 1269 825 537459 1610361 349875 0.0013 1.000 16.40 16.40 27.325 300 443 50 1336 900 596486 1784896 402175 0.0009 1.000 14.70 14.70 33.256 292 325 474 81 1440 975 694454 2073600 470000 0.0006 1.000 12.98 12.98 40.336 350 523 130 1572 1050 828830 2471184 552725 0.0005 1.000 11.27 11.27 44.951 375 575 182 1755 1125 1035803 3080025 661475 0.0004 1.000 9.94 9.94 52.199 400 657 264 1985 1200 1329283 3940225 798450 0.0003 1.000 8.62 8.62 59.727 425 753 360 2321 1275 998658 5387041 992775 -0.00001 1.000 7.44 7.44 450 911 518 2689 1335 1396930 7230721 1201475 0.00000 1.000 6.25 6.25 460 1025 632 1.000 5.89 Flame Front Position x Time to position x t (s) time from ignition ?t ?x ?(t 2 ) ?(t) 2 ?(t.x) V f F(t) " . . e q )(. " . . tFq e 1/?V f RIFT ? pynefloor particle board ? low preheat ? ISO ign data ? test 1 Ignition Time 75 m/s kW/m 2 (m/s) -0.5 10 75 0 33.82 20 77 2 231 60 17795 53361 4660 0.0050 0.319 32.76 10.46 14.142 30 79 4 237 90 18731 56169 7150 0.0050 0.324 31.96 10.34 14.142 40 81 6 250 120 20902 62500 10110 0.0016 0.328 29.56 9.68 24.985 50 90 15 271 15 24661 73441 1374 0.0011 0.345 27.16 9.38 30.836 60 100 25 300 180 30200 90000 18200 0.0010 0.364 24.97 9.09 31.623 70 10 35 335 21 37725 112225 2370 0.0008 0.382 22.98 8.77 35.590 80 125 50 380 240 48750 144400 30750 0.0006 0.407 20.99 8.54 41.975 90 45 70 44 27 65550 193600 4005 0.0004 0.438 18.76 8.22 47.532 100 170 95 510 300 87950 260100 51500 0.0004 0.475 16.53 7.85 50.000 110 95 120 593 33 118909 351649 6581 0.0003 0.508 14.65 7.45 54.022 120 228 153 698 360 165634 487204 84560 0.0002 0.550 13.12 7.21 63.568 130 75 200 821 39 228733 674041 107630 0.0002 0.604 11.59 7.00 67.104 140 318 243 966 420 315878 933156 136220 0.0002 0.649 10.50 6.81 70.175 150 73 298 1314 45 628382 1726596 200150 0.0001 0.703 9.40 6.61 131.635 160 623 548 1667 475 977499 2778889 266345 0.0000 0.909 8.39 7.63 145.962 165 671 596 0.943 7.94 293 Flame Front Position x Time to position x t (s) time from ignition ?t ?x ?(t 2 ) ?(t) 2 ?(t.x) V f F(t) " . . e q )(. " . . tFq e 1/?V f RIFT ? pynefloor particle board ? low preheat ? ISO ign data ? test 2 Ignition Time 65 m/s kW/m 2 (m/s) -0.5 10 65 0 33.82 20 68 3 204 60 13890 41616 4140 0.0033 0.300 32.76 9.83 17.321 30 71 6 213 90 15141 45369 6450 0.0033 0.307 31.96 9.80 17.321 40 74 9 226 120 17078 51076 9140 0.0019 0.313 29.56 9.26 22.949 50 81 16 249 15 20873 62001 12650 0.0010 0.328 27.16 8.90 32.094 60 94 29 27 180 26213 77841 1697 0.0009 0.353 24.97 8.81 34.008 70 104 39 320 21 34536 102400 22680 0.0007 0.371 22.98 8.53 37.922 80 22 57 368 240 45864 135424 2982 0.0005 0.402 20.99 8.44 43.609 90 142 77 449 27 69273 201601 41040 0.0003 0.434 18.76 8.14 57.358 100 85 120 521 300 92025 271441 5262 0.0003 0.495 16.53 8.19 54.502 110 194 129 608 33 124302 369664 67320 0.0004 0.507 14.65 7.43 49.559 120 229 164 703 360 168477 494209 8522 0.0002 0.551 13.12 7.23 65.952 130 80 215 842 39 241730 708964 110500 0.0002 0.609 11.59 7.06 72.115 140 333 268 1005 420 342953 1010025 141820 0.0002 0.664 10.50 6.97 74.869 150 392 327 0.721 9.40 294 Flame Front Position x Time to position x t (s) time from ignition ?t ?x ?(t 2 ) ?(t) 2 ?(t.x) V f F(t) " . . e q )(. " . . tFq e 1/?V f RIFT ? pynefloor particle board ? low preheat ? ISO ign data ? test 3 Ignition Time 71 m/s kW/m 2 (m/s) -0.5 10 71 0 0.307 33.82 20 76 5 227 60 17217 51529 4630 0.0022 0.317 32.76 10.40 21.257 30 80 9 244 90 19920 59536 7440 0.0016 0.326 31.96 10.41 24.944 40 88 17 272 120 24960 73984 11120 0.0008 0.341 29.56 10.09 35.277 50 104 33 310 150 32484 96100 15800 0.0007 0.371 27.16 10.08 38.759 60 118 47 362 18 44340 131044 2208 0.0005 0.395 24.97 9.87 42.774 70 40 69 419 210 59445 175561 29760 0.0005 0.431 22.98 9.90 46.372 80 161 90 483 24 78645 233289 3906 0.0005 0.462 20.99 9.69 45.826 90 82 111 555 270 103989 308025 50460 0.0004 0.491 18.76 9.21 50.759 100 212 141 640 30 138584 409600 6464 0.0003 0.530 16.53 8.76 56.605 110 46 175 752 330 191896 565504 83540 0.0002 0.571 14.65 8.37 64.342 120 294 223 101 36 369736 1024144 123700 0.0001 0.624 13.12 8.19 112.010 130 472 401 1280 390 573416 1638400 168600 0.0001 0.791 11.59 9.17 111.361 140 514 443 1572 42 830376 2471184 221220 0.0002 0.825 10.50 8.66 76.365 150 86 515 1749 441 1028793 3059001 257859 0.0001 0.881 9.40 8.28 109.870 151 649 578 0.927 9.30 295 Flame Front Position x Time to position x t (s) time from ignition ?t ?x ?(t 2 ) ?(t) 2 ?(t.x) V f F(t) " . . e q )(. " . . tFq e 1/?V f RIFT ? pynefloor particle board ? low preheat ? ISO ign data ? test 4 Ignition Time 80 m/s kW/m 2 (m/s) -0.5 10 80 0 0.326 33.82 20 82 2 249 60 20693 62001 5050 0.0027 0.330 32.76 10.80 19.272 30 87 7 258 90 22214 66564 7810 0.0027 0.340 31.96 10.85 19.272 40 89 9 273 120 24899 74529 11020 0.0018 0.343 29.56 10.15 23.664 50 97 17 295 15 29211 87025 1495 0.0010 0.359 27.16 9.74 31.833 60 109 29 329 180 36419 108241 20000 0.0008 0.380 24.97 9.49 36.091 70 23 43 373 21 46891 139129 2643 0.0006 0.404 22.98 9.28 40.104 80 141 61 427 240 61579 182329 34560 0.0005 0.432 20.99 9.07 44.796 90 63 83 499 27 84475 249001 4545 0.0004 0.465 18.76 8.72 52.258 100 195 115 579 300 113435 335241 58480 0.0003 0.508 16.53 8.40 53.948 110 221 141 680 33 156562 462400 7549 0.0003 0.541 14.65 7.93 59.328 120 64 184 789 360 210953 622521 95510 0.0002 0.591 13.12 7.76 64.435 130 304 224 1171 39 525721 1371241 155620 0.0000 0.635 11.59 7.36 142.295 140 603 523 155 420 870761 240560 220540 0.0000 0.894 10.50 9.38 142.348 150 44 564 1938 45 1255826 3755844 291580 0.0002 0.924 9.40 8.68 66.384 160 691 611 2145 474 1548317 4601025 340000 0.0001 0.957 8.39 8.03 115.901 164 810 730 1.000 8.03 296 Flame Front Position x Time to position x t (s) time from ignition ?t ?x ?(t 2 ) ?(t) 2 ?(t.x) V f F(t) " . . e q )(. " . . tFq e 1/?V f RIFT ? pynefloor particle board ? low preheat ? ISO ign data ? test 5 Ignition Time 82 m/s kW/m 2 (m/s) -0.5 10 82 0 0.330 33.82 20 85 3 258 60 22230 66564 5250 0.0021 0.336 32.76 10.99 21.602 30 91 9 270 90 24342 72900 8190 0.0021 0.347 31.96 11.10 21.602 40 94 12 283 120 26721 80089 11390 0.0028 0.353 29.56 10.43 18.772 50 98 16 30 15 30761 91809 1532 0.0011 0.360 27.16 9.79 30.486 60 111 29 335 180 37801 112225 20380 0.0007 0.384 24.97 9.58 37.448 70 26 44 379 21 48361 143641 2684 0.0006 0.409 22.98 9.39 39.377 80 142 60 428 240 61640 183184 34580 0.0006 0.434 20.99 9.10 41.255 90 60 78 493 27 82245 243049 4486 0.0004 0.460 18.76 8.64 50.075 100 191 109 571 300 110481 326041 57700 0.0003 0.503 16.53 8.32 54.782 110 220 138 670 33 151962 448900 7438 0.0003 0.540 14.65 7.91 58.519 120 259 177 789 360 211581 622521 95580 0.0002 0.586 13.12 7.69 67.281 130 310 228 938 390 299342 879844 123040 0.0002 0.641 11.59 7.43 74.227 140 369 287 1114 420 421486 1240996 157210 0.0002 0.699 10.50 7.34 79.098 150 435 353 1327 45 598915 1760929 200590 0.0001 0.759 9.40 7.14 88.048 160 523 441 1570 473 837298 2464900 248686 0.0001 0.833 8.39 6.99 116.745 163 612 530 0.901 8.12 Flame Front Position x Time to position x t (s) time from ignition ?t ?x ?(t 2 ) ?(t) 2 ?(t.x) V f F(t) " . . e q )(. " . . tFq e 1/?V f RIFT ? pynefloor particle board ? low preheat ? ISO ign data ? test 6 Ignition Time 78 m/s kW/m 2 (m/s) -0.5 10 78 0 0.322 33.82 297 20 81 2.5 242 60 19453 58322 4880 0.0040 0.327 32.76 10.70 15.811 30 83 5 252 90 21113 63252 7620 0.0026 0.332 31.96 10.60 19.720 40 88 10 266 120 23658 70756 10760 0.0017 0.341 29.56 10.09 24.608 50 95 17 285 15 27173 81225 1439 0.0014 0.355 27.16 9.64 26.458 60 102 24 310 180 32198 96100 18780 0.0011 0.368 24.97 9.18 30.246 70 13 35 335 21 37573 112225 2363 0.0011 0.387 22.98 8.89 30.246 80 120 42 367 240 45125 134689 29570 0.0009 0.399 20.99 8.37 32.998 90 34 56 410 27 56692 168100 3726 0.0005 0.421 18.76 7.90 42.774 100 156 78 475 300 76517 225625 48010 0.0004 0.455 16.53 7.52 50.656 110 85 107 563 33 107845 316969 6259 0.0003 0.495 14.65 7.26 57.586 120 222 144 666 360 150590 443556 80660 0.0003 0.542 13.12 7.12 60.828 130 59 181 790 39 211846 624100 103570 0.0002 0.586 11.59 6.79 66.200 140 309 231 94 420 300946 883600 132730 0.0002 0.640 10.50 6.72 75.332 150 372 294 0.702 9.40 Test 1 - half preheat Flame Front Position x Time to position x t (s) time from ignition ?t ?x ?(t 2 ) ?(t) 2 ?(t.x) V f F(t) " . . e q )(. " . . tFq e 1/?V f RIFT ? Pynefloor particle board ? half preheat ? ISO ign data ? test 1 Ignition Time 386 m/s kW/m 2 (m/s) -0.5 10 20 30 40 50 390 4 60 392 6 0.72 24.97 70 400 14 1199 210 479313 1437601 84080 0.0013 0.73 22.98 16.73 27.406 80 407 21 1217 240 493749 1481089 97460 0.0019 0.73 20.99 15.41 22.949 90 410 24 1231 270 505145 1515361 110860 0.0028 0.74 18.76 13.83 18.772 100 414 28 1251 300 521825 1565001 125270 0.0011 0.74 16.53 12.25 30.486 110 427 41 1286 330 551750 1653796 141770 0.0006 0.75 14.65 11.02 39.540 120 445 59 1348 360 606930 1817104 162250 0.0004 0.77 13.12 10.08 50.075 298 130 476 90 1432 390 685722 2050624 186820 0.0003 0.79 11.59 9.21 57.481 140 511 125 154 420 795722 237776 216670 0.0003 0.82 10.50 8.64 62.985 150 555 169 1680 450 946142 2822400 253030 0.0002 0.86 9.40 8.06 72.017 160 614 228 1834 480 1127246 3363556 294540 0.0002 0.90 8.39 7.57 74.227 170 665 279 204 510 1404446 417793 348990 0.0001 0.94 7.48 7.02 88.403 180 765 379 2275 540 1741475 5175625 411300 0.0001 1.00 6.57 6.57 95.063 190 845 459 1.00 5.93 5.93 Flame Front Position x Time to position x t (s) time from ignition ?t ?x ?(t 2 ) ?(t) 2 ?(t.x) V f F(t) " . . e q )(. " . . tFq e 1/?V f RIFT ? Pynefloor particle board ? half preheat ? ISO ign data ? test 2 Ignition Time 385 m/s kW/m 2 (m/s) -0.5 10 20 30 40 50 60 70 80 386 1 0.715 20.99 90 92 7 1178 270 462660 1387684 106160 0.0014 0.721 18.76 13.52 26.547 100 400 15 1212 300 490064 146894 121480 0.0007 0.728 16.53 12.04 38.545 110 420 35 1258 330 528244 1582564 138760 0.0005 0.746 14.65 10.93 43.609 120 438 53 1326 360 587268 1758276 159600 0.0004 0.762 13.12 10.00 49.497 130 468 83 1411 390 665893 1990921 184100 0.0003 0.788 11.59 9.13 57.984 140 505 120 1534 420 788770 2353156 215690 0.0002 0.818 10.50 8.59 68.664 150 561 176 1697 450 967907 2879809 255810 0.0002 0.862 9.40 8.10 79.536 160 631 246 1895 475 1207091 3591025 301105 0.0001 0.914 8.39 7.68 97.376 165 703 318 0.965 7.94 7.66 299 Flame Front Position x Time to position x t (s) time from ignition ?t ?x ?(t 2 ) ?(t) 2 ?(t.x) V f F(t) " . . e q )(. " . . tFq e 1/?V f RIFT ? Pynefloor particle board ? half preheat ? ISO ign data ? test 3 Ignition Time 386 m/s kW/m 2 (m/s) -0.5 10 20 30 40 50 60 70 80 389 3 0.718 20.99 90 93 7 1183 270 466571 1399489 106590 0.0016 0.722 18.76 13.54 24.944 100 401 15 1209 300 487475 1461681 121120 0.0009 0.729 16.53 12.05 33.575 110 415 29 1252 330 523122 1567504 138070 0.0006 0.742 14.65 10.87 42.111 120 436 50 1312 360 574842 1721344 157900 0.0004 0.760 13.12 9.98 48.019 130 461 75 1400 390 655626 1960000 182670 0.0003 0.782 11.59 9.06 58.497 140 503 117 1518 420 772446 2304324 213450 0.0002 0.816 10.50 8.57 68.297 150 554 168 1671 450 936921 2792241 251760 0.0002 0.857 9.40 8.05 74.580 160 614 228 1839 480 1134153 338192 295410 0.0002 0.902 8.39 7.57 76.494 170 671 285 2052 510 1415526 4210704 350370 0.0001 0.943 7.48 7.05 88.406 180 767 381 2323 538 1821755 5396329 418510 0.0001 1.000 6.57 6.57 109.437 188 885 499 1.000 6.06 6.06 Flame Front Position x Time to position x t (s) time from ignition ?t ?x ?(t 2 ) ?(t) 2 ?(t.x) V f F(t) " . . e q )(. " . . tFq e 1/?V f RIFT ? Pynefloor particle board ? half preheat ? ISO ign data ? test 4 Ignition Time 386 m/s kW/m 2 (m/s) -0.5 10 20 300 30 40 50 60 70 80 90 392 6 0.721 18.76 -6.82 100 401 15 1210 300 488354 1464100 121250 0.0008 0.729 16.53 12.05 35.814 110 417 31 1255 330 525659 1575025 138410 0.0006 0.743 14.65 10.89 42.514 120 437 51 1317 360 579227 1734489 158500 0.0004 0.761 13.12 9.99 48.094 130 463 77 1400 390 655338 1960000 182630 0.0003 0.783 11.59 9.08 56.409 140 500 114 151 420 763578 228010 212240 0.0002 0.814 10.50 8.54 64.960 150 547 161 1648 450 910410 2715904 248210 0.0002 0.851 9.40 8.00 71.120 160 601 215 1795 480 1079019 3222025 288200 0.0002 0.892 8.39 7.49 70.786 170 647 261 1986 510 1324454 3944196 338990 0.0001 0.926 7.48 6.93 84.240 180 738 352 2203 540 1632377 4853209 398250 0.0001 0.989 6.57 6.50 92.530 190 818 432 2461 565 2032793 6056521 464735 0.0001 1.000 5.93 5.93 105.792 195 905 519 1.000 5.61 5.61 Flame Front Position x Time to position x t (s) time from ignition ?t ?x ?(t 2 ) ?(t) 2 ?(t.x) V f F(t) " . . e q )(. " . . tFq e 1/?V f RIFT ? Pynefloor particle board ? half preheat ? ISO ign data ? test 5 Ignition Time 389 m/s kW/m 2 (m/s) -0.5 10 20 30 40 50 60 70 80 301 90 396 7 0.724 18.76 13.59 #DIV/0! 100 401 12 1209 300 487361 1461681 121060 0.0012 0.729 16.53 12.05 28.940 110 412 23 1242 330 514586 1542564 136900 0.0007 0.739 14.65 10.83 37.702 120 429 40 1300 360 564466 1690000 156470 0.0004 0.754 13.12 9.89 49.091 130 459 70 1377 390 633843 1896129 179610 0.0003 0.780 11.59 9.04 54.772 140 489 100 1479 420 731763 2187441 207780 0.0003 0.805 10.50 8.45 60.277 150 531 142 1612 450 871546 2598544 242830 0.0002 0.839 9.40 7.88 72.169 160 592 203 1774 480 1056226 3147076 285040 0.0002 0.886 8.39 7.43 77.463 170 651 262 1953 510 1278365 3814209 333190 0.0002 0.929 7.48 6.95 76.811 180 710 321 0.970 6.57 Flame Front Position x Time to position x t (s) time from ignition ?t ?x ?(t 2 ) ?(t) 2 ?(t.x) V f F(t) " . . e q )(. " . . tFq e 1/?V f RIFT ? Pynefloor particle board ? half preheat ? ISO ign data ? test 6 Ignition Time 387 m/s kW/m 2 (m/s) -0.5 10 20 30 40 50 60 70 80 389 -371 0.718 20.99 15.07 90 95 -365 1186 270 468950 1406596 106870 0.0015 0.723 18.76 13.57 25.520 100 402 -358 1215 300 492353 1476225 121730 0.0008 0.730 16.53 12.07 34.766 110 418 -342 1261 330 530809 1590121 139100 0.0005 0.744 14.65 10.91 44.395 120 441 -319 1332 360 592934 1774224 160390 0.0004 0.764 13.12 10.03 52.674 130 473 -287 1424 390 678310 2027776 185810 0.0003 0.792 11.59 9.18 58.788 302 140 510 -250 1536 420 789638 2359296 215840 0.0002 0.822 10.50 8.63 63.305 150 553 -207 1672 450 936790 2795584 251790 0.0002 0.856 9.40 8.05 70.558 160 609 -151 1835 480 1129619 3367225 294800 0.0002 0.898 8.39 7.54 77.517 170 673 -87 201 507 1361099 406022 341591 0.0001 0.944 7.48 7.06 85.339 177 733 -2 0.986 6.84 6.74 Flame Front Position x Time to positio n x t (s) time from ignition ?t ?x ?(t 2 ) ?(t) 2 ?(t.x) V f F(t) " . . e q )(. " . . tFq e 1/?V f RIFT ? Pynefloor particle board ? full preheat ? test 1 Ignition Time 760 m/s kW/m 2 (m/s) -0.5 10 20 30 40 50 60 70 80 90 778 18 1.00 18.76 100 785 25 2352 300 1844030 5531904 235310 0.0018 1.00 16.53 16.53 23.741 110 789 29 2374 33 1878746 5635876 26129 0.0012 1.00 14.65 14.65 28.363 120 800 40 2415 360 1944797 5832225 290170 0.0005 1.00 13.12 13.12 44.174 130 826 66 2482 39 2055012 6160324 32322 0.0004 1.00 11.59 11.59 52.960 140 85 96 2579 420 2219621 6651241 361770 0.0003 1.00 10.50 10.50 59.820 150 897 137 2725 45 2482129 7425625 40991 0.0002 1.00 9.40 9.40 77.240 160 972 212 2912 480 2837242 8479744 467380 0.0001 1.00 8.39 8.39 85.451 170 1043 283 3089 502 3186109 9541921 517558 0.0001 1.00 7.48 7.48 90.661 172 1074 314 1.00 7.30 - 303 Flame Front Position x Time to positio n x t (s) time from ignition ?t ?x ?(t 2 ) ?(t) 2 ?(t.x) V f F(t) " . . e q )(. " . . tFq e 1/?V f RIFT ? Pynefloor particle board ? full preheat ? test 2 Ignition Time 760s m/s kW/m 2 (m/s) -0.5 10 20 30 40 50 60 70 80 90 100 791 31 1.000 16.53 110 796 36 2394 330 1910546 5731236 263500 0.0012 1.000 14.65 14.65 28.940 120 807 47 2426 36 1962194 588547 29139 0.0007 1.000 13.12 13.12 36.952 130 823 63 2481 390 2052779 6155361 322970 0.0004 1.000 11.59 11.59 47.482 140 851 91 2563 42 2191851 6568969 35948 0.0003 1.000 10.50 10.50 57.665 150 889 129 2665 450 2370147 7102225 400490 0.0003 1.000 9.40 9.40 60.835 160 925 165 2804 48 2626046 7862416 44965 0.0002 1.000 8.39 8.39 72.033 170 990 230 2988 510 2987054 8928144 509440 0.0001 1.000 7.48 7.48 86.235 180 1073 313 1.000 6.57 Flame Front Position x Time to positio n x t (s) time from ignition ?t ?x ?(t 2 ) ?(t) 2 ?(t.x) V f F(t) " . . e q )(. " . . tFq e 1/?V f RIFT ? Pynefloor particle board ? full preheat ? test 3 Ignition Time 760 m/s kW/m 2 (m/s) -0.5 304 10 20 30 40 50 60 70 80 90 784 24 1.000 18.76 100 788 28 2365 300 1864449 5593225 236590 0.0022 1.000 16.53 16.53 21.257 110 793 33 2387 33 189942 5697769 26275 0.0010 1.000 14.65 14.65 30.972 120 806 46 2423 360 1957461 587092 291070 0.0006 1.000 13.12 13.12 39.540 130 824 64 2475 39 2042637 6125625 32214 0.0005 1.000 11.59 11.59 44.202 140 845 85 2554 420 2176226 6522916 358170 0.0003 1.000 10.50 10.50 56.113 150 88 125 2735 45 2507275 7480225 41185 0.0001 1.000 9.40 9.40 93.095 160 1005 245 2929 475 2872771 8579041 464985 0.0001 1.000 8.39 8.39 103.304 165 1039 279 1.000 7.94 Flame Front Position x Time to positio n x t (s) time from ignition ?t ?x ?(t 2 ) ?(t) 2 ?(t.x) V f F(t) " . . e q )(. " . . tFq e 1/?V f RIFT ? Pynefloor particle board ? full preheat ? test 4 Ignition Time 771 m/s kW/m 2 (m/s) -0.5 10 20 30 40 50 60 70 305 80 90 100 793 22 1.000 16.53 110 797 26 2399 330 1918539 5755201 264050 0.0012 1.000 14.65 14.65 29.439 120 809 38 2413 36 194093 5822569 28966 0.0012 1.000 13.12 13.12 28.752 130 807 36 2570 390 2215846 6604900 335550 0.0001 1.000 11.59 11.59 99.004 140 954 183 2750 420 2539486 7562500 386820 0.0001 1.000 10.50 10.50 101.236 150 989 218 2950 450 2902286 8702500 443030 0.0004 1.000 9.40 9.40 52.353 160 1007 236 1.000 8.39 Flame Front Position x Time to positio n x t (s) time from ignition ?t ?x ?(t 2 ) ?(t) 2 ?(t.x) V f F(t) " . . e q )(. " . . tFq e 1/?V f RIFT ? Pynefloor particle board ? full preheat ? test 5 Ignition Time 760 m/s kW/m 2 (m/s) -0.5 10 20 30 40 50 60 70 80 90 773 13 1.000 18.76 100 776 16 2329 300 1808105 5424241 232970 0.0028 1.000 16.53 16.53 18.772 110 780 20 2351 330 1842601 5527201 258800 0.0009 1.000 14.65 14.65 32.498 120 795 35 2294 36 1757386 5262436 27467 -0.0002 1.000 13.12 13.12 #NUM! 130 719 -41 2368 390 1878302 5607424 308430 0.0001 0.976 11.59 11.32 124.606 140 854 94 2464 42 2040158 6071296 34668 0.0001 1.000 10.50 10.50 97.625 306 150 891 131 2698 450 2431406 7279204 405690 0.0002 1.000 9.40 9.40 71.100 160 953 193 2857 479 2728259 8162449 457327 0.0002 1.000 8.39 8.39 80.124 169 1013 253 7.57 Flame Front Position x Time to positio n x t (s) time from ignition ?t ?x ?(t 2 ) ?(t) 2 ?(t.x) V f F(t) " . . e q )(. " . . tFq e 1/?V f RIFT ? Pynefloor particle board ? full preheat ? test 6 Ignition Time 760 m/s kW/m 2 (m/s) -0.5 10 20 30 40 50 60 70 80 90 772 12 1.000 18.76 100 779 19 2339 300 1823769 5470921 234060 0.0012 1.000 16.53 16.53 28.358 110 788 28 2379 33 188712 565964 26202 0.0006 1.000 14.65 14.65 41.996 120 812 52 2438 360 1982532 5943844 293060 0.0004 1.000 13.12 13.12 50.013 130 838 78 2520 39 2118488 6350400 32818 0.0003 1.000 11.59 11.59 53.948 140 870 110 2628 420 2305544 6906384 368740 0.0002 1.000 10.50 10.50 64.543 150 92 16 2758 45 254032 760656 41468 0.0002 1.000 9.40 9.40 70.005 160 968 208 2969 479 2951985 8814961 475569 0.0001 1.000 8.39 8.39 94.857 169 1081 321 1.000 7.57 307 Superflake particle board Flame Front Position x Time to position t time from ignition ?t ?x ?(t 2 ) ?(t) 2 ?(t.x) V F(t) q? q?*F(t) 1/?V RIFT ? Superflake PB ? low preheat ? ISO ign data ? test 1 Ignition Time 82 m/s kW/m 2 (m/s) -0.5 10 33.82 20 32.76 30 89 7 31.96 40 95 13 288 120 27762 82944 11670 0.0013 0.319 29.56 9.42 27.568 50 104 22 318 150 34002 101124 16140 0.0008 0.334 27.16 9.06 35.000 60 19 37 359 180 43473 128881 21860 0.0006 0.357 24.97 8.91 40.026 70 136 54 430 210 63282 184900 30660 0.0003 0.381 22.98 8.76 54.259 80 75 93 493 240 82245 243049 39900 0.0004 0.433 20.99 9.08 51.682 90 182 100 571 270 109545 326041 51780 0.0005 0.441 18.76 8.28 47.086 100 214 132 642 300 139436 412164 64840 0.0003 0.479 16.53 7.91 56.569 110 46 164 757 330 194521 573049 84100 0.0002 0.513 14.65 7.52 64.981 120 297 215 890 359 269134 792100 107463 0.0002 0.564 13.12 7.40 72.904 129 347 265 0.609 11.75 7.16 Flame Front Position x Time to position x t (s) time from ignition ?t ?x ?(t 2 ) ?(t) 2 ?(t.x) V f F(t) " . . e q )(. " . . tFq e 1/?V f RIFT ? Superflake PB ? low preheat ? ISO ign data ? test 2 Ignition Time 82 m/s kW/m 2 (m/s) -0.5 34.88 10 33.82 20 32.76 30 82 31.96 40 93 11 29.56 50 100 18 309 150 32105 95481 15680 0.0008 0.327 27.16 8.88 34.766 60 16 34 343 180 3958 117649 20850 0.0007 0.352 24.97 8.80 36.952 308 70 127 45 393 210 52085 154449 27850 0.0006 0.369 22.98 8.47 42.078 80 50 68 447 240 67529 199809 36190 0.0005 0.401 20.99 8.41 46.406 90 170 88 518 270 90604 268324 47100 0.0004 0.426 18.76 8.00 49.216 100 98 116 597 300 120545 356409 60290 0.0003 0.460 16.53 7.61 54.337 110 229 147 707 330 170045 499849 78590 0.0002 0.495 14.65 7.25 64.663 120 80 198 840 360 240402 705600 101820 0.0002 0.547 13.12 7.18 71.414 130 331 249 994 390 334650 988036 130250 0.0002 0.595 11.59 6.90 71.765 140 83 301 1177 420 470619 1385329 166100 0.0001 0.640 10.50 6.72 81.847 150 463 381 1424 450 695142 2027776 215550 0.0001 0.704 9.40 6.62 99.271 160 578 496 1748 320 548453 3055504 161930 0.0001 0.786 8.39 6.60 138.446 167 707 625 0.870 7.76 6.74 Flame Front Position x Time to position x t (s) time from ignition ?t ?x ?(t 2 ) ?(t) 2 ?(t.x) V f F(t) " . . e q )(. " . . tFq e 1/?V f RIFT ? Superflake PB ? low preheat ? ISO ign data ? test 3 Ignition Time 83 m/s kW/m 2 (m/s) -0.5 10 33.82 20 32.76 30 83 0 31.96 40 91 8 29.56 50 7 14 302 150 30686 91204 15330 0.0008 0.322 27.16 8.75 35.181 60 114 31 342 180 3956 116964 20860 0.0006 0.349 24.97 8.72 41.231 70 31 48 394 210 52358 155236 27930 0.0006 0.374 22.98 8.60 41.839 80 149 66 455 240 69987 207025 36840 0.0004 0.399 20.99 8.38 47.162 90 75 92 530 270 95262 280900 48270 0.0003 0.433 18.76 8.12 53.454 100 206 123 629 300 134565 395641 63630 0.0003 0.469 16.53 7.76 60.643 110 48 165 740 330 185736 547600 82200 0.0002 0.515 14.65 7.55 63.272 120 286 203 869 360 255525 755161 105150 0.0002 0.553 13.12 7.26 66.130 130 335 252 1053 390 380645 1108809 138350 0.0001 0.599 11.59 6.94 86.966 140 432 349 1257 420 538949 158004 177530 0.0001 0.680 10.50 7.14 88.958 150 90 407 1507 442 768949 2271049 222900 0.0001 0.724 9.40 6.81 117.204 309 152 585 502 0.791 9.20 7.28 Flame Front Position x Time to position x t (s) time from ignition ?t ?x ?(t 2 ) ?(t) 2 ?(t.x) V f F(t) " . . e q )(. " . . tFq e 1/?V f RIFT ? Superflake PB ? low preheat ? ISO ign data ? test 4 Ignition Time 82 m/s kW/m 2 (m/s) -0.5 10 33.82 20 32.76 30 82 0 31.96 40 91 9 272 120 24806 73984 11050 0.0012 0.312 29.56 9.22 29.172 50 9 17 300 150 30182 90000 15190 0.0010 0.325 27.16 8.84 30.950 60 110 28 335 180 37777 112225 20370 0.0007 0.343 24.97 8.56 36.952 70 26 44 381 210 49001 145161 27020 0.0006 0.367 22.98 8.44 41.884 80 145 63 459 240 72245 210681 37340 0.0003 0.394 20.99 8.27 57.051 90 88 106 528 270 94394 278784 48020 0.0003 0.448 18.76 8.41 54.148 100 195 113 616 300 127658 379456 62050 0.0004 0.457 16.53 7.55 51.048 110 233 151 695 330 163603 483025 77170 0.0003 0.499 14.65 7.32 60.031 120 67 185 825 360 231203 680625 99920 0.0002 0.534 13.12 7.01 68.588 130 325 243 1057 390 393139 1117249 139390 0.0001 0.590 11.59 6.84 102.303 140 465 383 1309 420 591211 1713481 185200 0.0001 0.705 10.50 7.40 101.66 150 519 437 1579 443 839611 249324 233985 0.0001 0.745 9.40 7.00 102.038 153 95 513 0.798 9.10 7.26 Flame Front Position x Time to position x t (s) time from ignition ?t ?x ?(t 2 ) ?(t) 2 ?(t.x) V f F(t) " . . e q )(. " . . tFq e 1/?V f RIFT ? Superflake PB ? low preheat ? ISO ign data ? test 5 Ignition Time 82 m/s kW/m 2 (m/s) -0.5 10 33.82 20 82 32.76 30 8 6 262 90 22932 68644 7960 0.0020 0.307 31.96 9.81 22.509 40 92 10 282 120 2661 79524 11420 0.0013 0.314 29.56 9.27 27.255 310 50 102 20 311 150 32557 96721 15800 0.0008 0.330 27.16 8.97 35.590 60 17 35 355 180 42589 126025 21640 0.0006 0.354 24.97 8.83 41.326 70 136 54 409 210 56521 167281 29020 0.0005 0.381 22.98 8.76 44.164 80 56 74 475 240 7632 225625 38470 0.0004 0.409 20.99 8.57 48.656 90 183 101 559 270 106225 312481 50950 0.0003 0.442 18.76 8.30 56.798 100 220 138 652 300 143890 425104 65860 0.0003 0.485 16.53 8.02 57.586 110 49 167 772 330 202210 595984 85750 0.0002 0.516 14.65 7.56 65.387 120 303 221 907 360 279835 822649 109900 0.0002 0.569 13.12 7.47 72.805 130 55 273 1079 390 395075 1164241 141450 0.0002 0.616 11.59 7.14 76.991 140 421 339 1285 420 562347 1651225 181440 0.0001 0.671 10.50 7.04 88.048 150 509 427 1486 443 745458 2208196 220358 0.0001 0.738 9.40 6.94 100.750 153 56 474 0.771 9.10 7.02 Flame Front Position x Time to position x t (s) time from ignition ?t ?x ?(t 2 ) ?(t) 2 ?(t.x) V f F(t) " . . e q )(. " . . tFq e 1/?V f RIFT ? Superflake PB ? low preheat ? ISO ign data ? test 6 Ignition Time 82 m/s kW/m 2 (m/s) -0.5 10 33.82 20 32.76 30 82 0 175 90 15373 30625 6180 0.0002 0.296 31.96 9.47 74.521 40 93 11 283 120 27037 80089 11580 0.0008 0.315 29.56 9.32 36.197 50 108 26 324 150 35442 104976 16500 0.0007 0.340 27.16 9.23 38.730 60 23 41 368 180 4556 135424 22370 0.0007 0.363 24.97 9.06 38.086 70 137 55 422 210 60142 178084 29930 0.0005 0.383 22.98 8.80 44.740 80 62 80 492 240 8226 242064 39920 0.0004 0.416 20.99 8.74 53.016 90 193 111 582 270 115022 338724 53030 0.0003 0.454 18.76 8.52 57.029 100 227 145 689 300 161139 474721 69660 0.0003 0.493 16.53 8.15 61.758 110 69 187 816 330 226290 665856 90690 0.0002 0.536 14.65 7.86 68.297 120 320 238 939 360 297261 881721 113490 0.0002 0.585 13.12 7.68 64.349 130 50 268 0.612 11.59 7.09 311 Flame Front Position x Time to position x t (s) time from ignition ?t ?x ?(t 2 ) ?(t) 2 ?(t.x) V f F(t) " . . e q )(. " . . tFq e 1/?V f RIFT ? Superflake Particle Board Full preheat ? test 1 Ignition Time 699 m/s kW/m 2 (m/s) -0.5 10 20 30 40 50 60 70 80 90 100 706 7 0.869 16.53 110 711 12 2139 330 1525241 4575321 235450 0.0012 0.872 14.65 12.78 28.940 120 722 23 2175 360 1577369 4730625 261310 0.0006 0.879 13.12 11.53 39.919 130 742 43 2235 390 1666289 4995225 291040 0.0004 1.000 11.59 11.59 49.775 140 771 72 2322 420 1799486 5391684 325750 0.0003 1.000 10.50 10.50 58.053 150 809 110 2432 450 1974826 5914624 365610 0.0002 1.000 9.40 9.40 63.680 160 852 153 2587 478 2237861 6692569 413238 0.0001 1.000 8.39 8.39 81.964 168 926 227 1.000 7.66 Flame Front Position x Time to position x t (s) time from ignition ?t ?x ?(t 2 ) ?(t) 2 ?(t.x) V f F(t) " . . e q )(. " . . tFq e 1/?V f RIFT ? Superflake Particle Board Full preheat ? test 2 Ignition Time 755 m/s kW/m 2 (m/s) -0.5 10 20 30 312 40 50 60 70 80 90 100 763 8 16.53 110 771 16 2318 330 1791266 5373124 255190 0.0009 1.000 14.65 14.65 32.708 120 784 29 2357 360 1852301 5555449 283150 0.0006 1.000 13.12 13.12 39.540 130 802 47 2420 390 1953416 5856400 315100 0.0004 1.000 11.59 11.59 50.649 140 834 79 2514 420 2109644 6320196 352720 0.0003 1.000 10.50 10.50 61.900 150 878 123 2648 450 2342536 7011904 398220 0.0002 1.000 9.40 9.40 71.638 160 936 181 2814 320 1646980 7918596 281460 0.0000 1.000 8.39 8.39 230.386 170 1000 245 2990 505 2987012 8940100 504210 0.0001 1.000 7.48 7.48 88.385 175 1054 299 1.000 7.02 Flame Front Position x Time to position x t (s) time from ignition ?t ?x ?(t 2 ) ?(t) 2 ?(t.x) V f F(t) " . . e q )(. " . . tFq e 1/?V f RIFT ? Superflake Particle Board Full preheat ? test 3 Ignition Time 755 m/s kW/m 2 (m/s) -0.5 10 20 30 40 50 60 70 80 90 100 761 6 1.000 16.53 110 769 14 2312 330 1782006 5345344 254530 0.0009 1.000 14.65 14.65 32.708 313 120 782 27 2351 360 1842885 5527201 282430 0.0006 1.000 13.12 13.12 39.540 130 800 45 2409 390 1935453 5803281 313620 0.0004 1.000 11.59 11.59 47.749 140 827 72 2493 420 2073885 6215049 349680 0.0003 1.000 10.50 10.50 57.761 150 866 111 2595 450 2247489 6734025 390000 0.0003 1.000 9.40 9.40 61.254 160 902 147 2783 479 2593785 7745089 445755 0.0001 1.000 8.39 8.39 92.835 169 1015 260 1.000 7.57 Flame Front Position x Time to position x t (s) time from ignition ?t ?x ?(t 2 ) ?(t) 2 ?(t.x) V f F(t) " . . e q )(. " . . tFq e 1/?V f RIFT ? Superflake Particle Board Full preheat ? test4 Ignition Time 755 m/s kW/m 2 (m/s) -0.5 10 20 30 40 50 60 70 80 90 100 762 7 1.000 16.53 110 769 14 2314 330 1785094 5354596 254750 0.0009 1.000 14.65 14.65 32.998 120 783 28 2368 360 1870306 5607424 284630 0.0004 1.000 13.12 13.12 49.780 130 816 61 2451 390 2004849 6007401 319320 0.0003 1.000 11.59 11.59 58.755 140 852 97 2565 420 2196369 6579225 359910 0.0002 1.000 10.50 10.50 63.770 150 897 142 2714 448 2461738 7365796 406300 0.0002 1.000 9.40 9.40 80.080 158 965 210 1.000 8.60 314 Flame Front Position x Time to position x t (s) time from ignition ?t ?x ?(t 2 ) ?(t) 2 ?(t.x) V f F(t) " . . e q )(. " . . tFq e 1/?V f RIFT ? Superflake Particle Board Full preheat ? test 5 Ignition Time 756 m/s kW/m 2 (m/s) -0.5 10 20 30 40 50 60 70 80 90 100 765 9 1535 300 1178125 2356225 161200 0.0000 1.000 16.53 16.53 225.837 110 770 14 2321 330 1795921 5387041 255520 0.0009 1.000 14.65 14.65 33.853 120 786 30 2362 360 1860332 5579044 283800 0.0006 1.000 13.12 13.12 42.514 130 806 50 2424 390 1959656 5875776 315580 0.0004 1.000 11.59 11.59 48.094 140 832 76 2507 420 2097021 6285049 351610 0.0003 1.000 10.50 10.50 56.409 150 869 113 2619 450 2290109 6859161 393710 0.0002 1.000 9.40 9.40 65.787 160 918 162 2770 480 2564174 7672900 444340 0.0002 1.000 8.39 8.39 75.746 170 983 227 2933 505 2874037 8602489 494590 0.0001 1.000 7.48 7.48 86.790 175 1032 276 1.000 7.02 Flame Front Position x Time to position x t (s) time from ignition ?t ?x ?(t 2 ) ?(t) 2 ?(t.x) V f F(t) " . . e q )(. " . . tFq e 1/?V f RIFT ? Superflake Particle Board Full preheat ? test 6 Ignition Time 755 m/s kW/m 2 (m/s) -0.5 10 20 30 40 315 50 60 70 80 90 100 765 10 1.000 16.53 110 771 16 2320 330 1794322 5382400 255390 0.0010 1.000 14.65 14.65 31.512 120 784 29 2358 360 1853906 5560164 283280 0.0006 1.000 13.12 13.12 40.234 130 803 48 2418 390 1950026 5846724 314810 0.0004 1.000 11.59 11.59 48.772 140 831 76 2501 420 2087059 6255001 350780 0.0003 1.000 10.50 10.50 56.716 150 867 112 2609 450 2272171 6806881 392150 0.0002 1.000 9.40 9.40 63.351 160 911 156 2763 480 2551835 7634169 443260 0.0002 1.000 8.39 8.39 77.635 170 985 230 2956 510 2923746 8737936 504010 0.0001 1.000 7.48 7.48 86.314 180 1060 305 1.000 6.57 316 Radiata Pine Flame Front Position x Time to position x t (s) time from ignition ?t ?x ?(t 2 ) ?(t) 2 ?(t.x) V f F(t) " . . e q )(. " . . tFq e 1/?V f LIFT ? Radiata Pine test 1 Ignition Time 635 m/s kW/m 2 (m/s) -0.5 50 7 100 125 150 175 200 225 250 275 300 653 18 1.000 16.50 325 656 21 1972 975 1296314 3888784 641150 0.0047 1.000 14.90 14.90 14.514 350 663 28 1997 1050 1329589 3988009 699500 0.0022 1.000 13.20 13.20 21.433 375 678 43 2038 1125 1385062 4153444 765100 0.0015 1.000 11.50 11.50 26.137 400 697 62 2094 1200 1462454 4384836 838625 0.0012 1.000 9.80 9.80 28.661 425 719 84 2168 1275 156827 4700224 922775 0.0009 1.000 8.40 8.40 33.387 450 752 117 2261 1350 1706565 5112121 1019225 0.0007 1.000 6.90 6.90 37.714 475 790 155 2380 1425 1891848 5664400 1132650 0.0006 1.000 6.00 6.00 41.566 500 838 203 2493 1480 2074569 6215049 1231075 0.0004 1.000 5.00 5.00 49.143 505 865 230 4.84 Flame Front Position x Time to position x t (s) time from ignition ?t ?x ?(t 2 ) ?(t) 2 ?(t.x) V f F(t) " . . e q )(. " . . tFq e 1/?V f LIFT ? Radiata Pine test 2 Ignition Time 639 m/s kW/m 2 (m/s) -0.5 50 7 317 100 125 150 175 200 225 250 275 300 657 18 1.000 16.50 325 669 30 2001 975 1334835 4004001 650775 0.0027 1.000 14.90 14.90 19.322 350 675 36 2039 1050 1386211 4157521 714300 0.0018 1.000 13.20 13.20 23.880 375 69 56 208 1125 145561 4363921 784475 0.0011 1.000 11.50 11.50 29.706 400 719 80 2163 1200 1560987 4678569 866550 0.0009 1.000 9.80 9.80 32.931 425 74 110 2262 1275 1708398 5116644 963225 0.0007 1.000 8.40 8.40 38.987 450 794 155 2383 1350 1897037 5678689 1074625 0.0005 1.000 6.90 6.90 42.662 475 840 201 2601 1425 2271125 6765201 1239800 0.0003 1.000 6.00 6.00 60.933 500 967 328 1.000 5.00 Flame Front Position x Time to position x t (s) time from ignition ?t ?x ?(t 2 ) ?(t) 2 ?(t.x) V f F(t) " . . e q )(. " . . tFq e 1/?V f LIFT ? Radiata Pine test 3 Ignition Time 655 m/s kW/m 2 (m/s) -0.5 50 7 100 125 150 175 200 225 250 275 318 300 325 350 675 20 13.20 375 682 27 2051 1125 1402385 4206601 769600 0.0026 1.000 11.50 11.50 19.717 400 694 39 2090 1200 1456556 4368100 836800 0.0015 1.000 9.80 9.80 25.560 425 71 59 2147 1275 1537553 4609609 913600 0.0011 1.000 8.40 8.40 30.062 450 739 84 2228 1350 1656542 4963984 1004125 0.0008 1.000 6.90 6.90 35.117 475 775 120 2382 1425 1900170 5673924 113467 0.0004 1.000 6.00 6.00 52.420 500 868 213 2608 1500 2285274 6801664 1308750 0.0003 1.000 5.00 5.00 61.649 525 965 310 286 1564 275587 8225424 149849 0.0002 1.000 4.20 4.20 65.228 539 1035 380 3.75 319 Flame Front Position x Time to position x t (s) time from ignition ?t ?x ?(t 2 ) ?(t) 2 ?(t.x) V f F(t) " . . e q )(. " . . tFq e 1/?V f RIFT ? Pine- low preheat ISO ign data- test 1 Ignition Time 43 m/s kW/m 2 (m/s) -0.5 10 42 0 0.256 33.88 20 43 1 129 60 5549 16641 2600 0.0100 0.259 32.87 8.50 10.000 30 44 2 132 90 5810 17424 3980 0.0100 0.262 32.16 8.41 10.000 40 45 3 136 120 6170 18496 5470 0.0064 0.264 29.67 7.85 12.472 50 47 5 145 150 7043 21025 7330 0.0023 0.270 27.17 7.34 20.817 60 53 11 161 180 8739 25921 9800 0.0014 0.287 24.95 7.16 26.547 70 61 19 184 210 11430 33856 13050 0.0012 0.308 22.99 7.08 29.172 80 70 28 213 240 15345 45369 17250 0.0009 0.330 21.03 6.94 32.514 90 82 40 247 270 20649 61009 22480 0.0008 0.357 18.88 6.74 35.365 100 95 53 297 300 30149 88209 30080 0.0005 0.384 16.73 6.43 44.308 110 120 78 363 330 45329 131769 40460 0.0004 0.432 15.06 6.50 51.506 120 148 106 440 360 65888 193600 53320 0.0004 0.480 13.89 6.66 51.040 130 172 130 521 390 91889 271441 68260 0.0004 0.517 12.71 6.57 51.554 140 201 159 603 420 122885 363609 85000 0.0003 0.559 11.59 6.48 53.852 150 230 188 688 450 159350 473344 103760 0.0004 0.598 10.48 6.26 52.926 160 257 215 773 480 200745 597529 124240 0.0004 0.632 9.39 5.93 52.926 170 286 244 862 510 249606 743044 147160 0.0003 0.667 8.33 5.56 55.716 180 319 277 959 540 308873 919681 173300 0.0003 0.704 7.27 5.12 58.318 190 354 312 1079 570 391913 1164241 205880 0.0002 0.742 6.55 4.86 66.373 200 406 364 0.794 5.83 4.63 320 Flame Front Position x Time to position x t (s) time from ignition ?t ?x ?(t 2 ) ?(t) 2 ?(t.x) V f F(t) " . . e q )(. " . . tFq e 1/?V f RIFT ? Radiata Pine- low preheat ISO ign data- test 2 Ignition Time 42 m/s kW/m 2 (m/s) -0.5 10 43 0 0.259 33.88 20 44 1 133 60 5901 17689 2690 0.0064 0.262 32.87 8.60 12.472 30 46 3 139 90 6453 19321 4220 0.0039 0.267 32.16 8.60 15.916 40 49 6 148 120 7326 21904 5990 0.0028 0.276 29.67 8.19 18.772 50 53 10 161 150 8691 25921 8150 0.0020 0.287 27.17 7.80 22.509 60 59 16 180 180 10914 32400 10950 0.0013 0.303 24.95 7.56 27.568 70 68 25 203 210 13881 41209 14380 0.0012 0.325 22.99 7.48 29.172 80 76 33 230 240 17796 52900 18580 0.0011 0.344 21.03 7.23 30.062 90 86 43 258 270 22388 66564 23420 0.0010 0.366 18.88 6.90 31.623 100 96 53 294 300 29156 86436 29660 0.0008 0.386 16.73 6.46 36.374 110 112 69 335 330 37889 112225 37160 0.0006 0.417 15.06 6.28 39.377 120 127 84 387 360 50577 149769 46800 0.0006 0.444 13.89 6.17 42.622 130 148 105 445 390 66933 198025 58280 0.0005 0.480 12.71 6.10 46.372 140 170 127 517 420 90405 267289 72890 0.0004 0.514 11.59 5.96 50.656 150 199 156 601 450 122325 361201 90770 0.0003 0.556 10.48 5.83 55.716 160 232 189 697 480 164181 485809 112190 0.0003 0.601 9.39 5.64 57.881 170 266 223 808 504 220680 652864 136280 0.0002 0.643 8.33 5.36 75.541 174 310 267 0.694 7.91 Flame Front Position x Time to position x t (s) time from ignition ?t ?x ?(t 2 ) ?(t) 2 ?(t.x) V f F(t) " . . e q )(. " . . tFq e 1/?V f RIFT ? Radiata Pine- low preheat ISO ign data- test 3 Ignition Time 40 m/s kW/m 2 (m/s) -0.5 10 40 0 0.249 33.88 20 41 1 123 60 5045 15129 2480 0.0100 0.252 32.87 8.30 10.000 321 30 42 2 126 90 5294 15876 3800 0.0100 0.256 32.16 8.22 10.000 40 43 3 133 120 5917 17689 5380 0.0029 0.259 29.67 7.67 18.559 50 48 8 145 150 7069 21025 7360 0.0018 0.273 27.17 7.42 23.484 60 54 14 163 180 8941 26569 9910 0.0015 0.290 24.95 7.23 25.520 70 61 21 186 210 11678 34596 13190 0.0012 0.308 22.99 7.08 29.306 80 71 31 220 240 16506 48400 17870 0.0007 0.332 21.03 6.99 37.152 90 88 48 259 270 22785 67081 23600 0.0007 0.370 18.88 6.98 38.267 100 100 60 299 300 30065 89401 30130 0.0009 0.394 16.73 6.59 33.922 110 111 71 337 330 38197 113569 37330 0.0008 0.415 15.06 6.26 36.197 120 126 86 411 360 58473 168921 49950 0.0003 0.443 13.89 6.14 58.635 130 174 134 484 390 80008 234256 63500 0.0003 0.520 12.71 6.61 57.576 140 184 144 570 420 109076 324900 80180 0.0005 0.535 11.59 6.20 45.190 150 212 172 636 450 136400 404496 95960 0.0004 0.574 10.48 6.01 52.915 160 240 200 726 480 177620 527076 116780 0.0003 0.611 9.39 5.73 55.764 170 274 234 833 505 234437 693889 140805 0.0002 0.653 8.33 5.44 73.376 175 319 279 0.704 7.80 Flame Front Position x Time to position x t (s) time from ignition ?t ?x ?(t 2 ) ?(t) 2 ?(t.x) V f F(t) " . . e q )(. " . . tFq e 1/?V f RIFT ? Pine- low preheat ISO ign data- test 4 Ignition Time 40 m/s kW/m 2 (m/s) -0.5 10 40 0 0.249 33.88 20 41 1 123 60 5045 15129 2480 0.0100 0.252 32.87 8.30 10.000 30 42 2 129 90 5561 16641 3920 0.0036 0.256 32.16 8.22 16.733 40 46 6 140 120 6584 19600 5700 0.0020 0.267 29.67 7.93 22.509 50 52 12 153 150 7845 23409 7740 0.0021 0.284 27.17 7.73 21.602 60 55 15 166 180 9210 27556 10030 0.0028 0.292 24.95 7.29 18.772 70 59 19 182 210 11130 33124 12870 0.0015 0.303 22.99 6.96 26.116 80 68 28 203 240 13881 41209 16410 0.0012 0.325 21.03 6.84 29.172 322 90 76 36 229 270 17625 52441 20780 0.0012 0.344 18.88 6.49 29.172 100 85 45 258 300 22410 66564 26010 0.0009 0.363 16.73 6.08 32.514 110 97 57 292 330 28734 85264 32370 0.0008 0.388 15.06 5.85 35.365 120 110 70 333 360 37385 110889 40250 0.0007 0.414 13.89 5.74 38.147 130 126 86 381 390 49001 145161 49880 0.0006 0.443 12.71 5.62 41.884 140 145 105 446 420 67526 198916 62930 0.0004 0.475 11.59 5.50 49.911 150 175 135 559 450 108771 312481 84790 0.0002 0.522 10.48 5.46 70.035 160 239 199 680 480 158502 462400 109710 0.0002 0.610 9.39 5.72 69.287 170 266 226 804 510 217278 646416 137280 0.0003 0.643 8.33 5.36 54.863 180 299 259 906 538 276438 820836 163148 0.0002 0.682 7.27 4.96 64.849 188 341 301 0.728 6.70 Flame Front Position x Time to position x t (s) time from ignition ?t ?x ?(t 2 ) ?(t) 2 ?(t.x) V f F(t) " . . e q )(. " . . tFq e 1/?V f RIFT ? Radiata Pine- low preheat ISO ign data- test 5 Ignition Time 40 m/s kW/m 2 (m/s) -0.5 10 40 0 0.249 33.88 20 42 2 125 60 5213 15625 2530 0.0064 0.256 32.87 8.40 12.472 30 43 3 132 90 5822 17424 4010 0.0036 0.259 32.16 8.31 16.733 40 47 7 143 120 6867 20449 5820 0.0020 0.270 29.67 8.02 22.509 50 53 13 163 150 8987 26569 8310 0.0012 0.287 27.17 7.80 28.577 60 63 23 190 180 12254 36100 11610 0.0010 0.313 24.95 7.81 32.416 70 74 34 219 210 16169 47961 15520 0.0010 0.339 22.99 7.80 30.950 80 82 42 247 240 20481 61009 19930 0.0012 0.357 21.03 7.51 29.172 90 91 51 279 270 26241 77841 25350 0.0008 0.376 18.88 7.10 35.000 100 106 66 323 300 35393 104329 32650 0.0006 0.406 16.73 6.79 41.975 110 126 86 382 330 49612 145924 42460 0.0005 0.443 15.06 6.67 46.969 120 150 110 450 360 68652 202500 54480 0.0004 0.483 13.89 6.70 48.990 130 174 134 532 390 96040 283024 69740 0.0003 0.520 12.71 6.61 54.118 140 208 168 635 420 137549 403225 89690 0.0003 0.569 11.59 6.59 63.052 150 253 213 745 450 187929 555025 112510 0.0003 0.627 10.48 6.57 61.992 323 160 284 244 864 480 251594 746496 138980 0.0003 0.664 9.39 6.24 61.094 170 327 287 0.713 8.33 Flame Front Position x Time to position x t (s) time from ignition ?t ?x ?(t 2 ) ?(t) 2 ?(t.x) V f F(t) " . . e q )(. " . . tFq e 1/?V f RIFT ? Radiata Pine- low preheat ISO ign data- test 6 Ignition Time 40 m/s kW/m 2 (m/s) -0.5 10 40 0 0.249 33.88 20 43 3 129 60 5565 16641 2640 0.0033 0.259 32.87 8.50 17.321 30 46 6 141 90 6669 19881 4320 0.0021 0.267 32.16 8.60 21.602 40 52 12 157 120 8301 24649 6410 0.0015 0.284 29.67 8.43 25.520 50 59 19 178 150 10674 31684 9050 0.0013 0.303 27.17 8.23 27.406 60 67 27 209 180 14859 43681 12780 0.0008 0.323 24.95 8.05 35.277 70 83 43 255 210 22403 65025 18230 0.0005 0.359 22.99 8.26 43.770 80 105 65 315 240 34043 99225 25640 0.0005 0.404 21.03 8.50 46.904 90 127 87 377 270 48179 142129 34330 0.0005 0.444 18.88 8.39 44.796 100 145 105 441 300 65715 194481 44520 0.0005 0.475 16.73 7.94 45.981 110 169 129 508 330 87222 258064 56370 0.0004 0.513 15.06 7.72 49.501 120 194 154 585 360 115481 342225 70730 0.0004 0.549 13.89 7.62 51.506 130 222 182 674 390 153484 454276 88260 0.0003 0.587 12.71 7.47 56.716 140 258 218 802 420 219532 643204 113280 0.0002 0.633 11.59 7.34 71.629 150 322 282 958 450 313132 917764 144900 0.0002 0.707 10.48 7.41 77.517 160 378 338 1203 480 499577 1447209 194290 0.0001 0.767 9.39 7.20 97.408 170 503 463 0.884 8.33 Flame Front Position x Time to position x t (s) time from ignition ?t ?x ?(t 2 ) ?(t) 2 ?(t.x) V f F(t) " . . e q )(. " . . tFq e 1/?V f RIFT ? Radiata Pine ? full preheat ? test 1 Ignition Time 643 m/s kW/m 2 (m/s) -0.5 10 324 20 30 40 50 60 70 80 90 100 110 652 9 1.000 15.06 120 657 14 1969 360 1292353 3876961 236360 0.0024 1.000 13.89 13.89 20.207 130 660 17 1984 390 1312138 3936256 258020 0.0019 1.000 12.71 12.71 22.949 140 667 24 2002 420 1336114 4008004 280430 0.0013 1.000 11.59 11.59 27.406 150 675 32 2025 450 1367003 4100625 303910 0.0013 1.000 10.48 10.48 28.284 160 683 40 2051 480 1402363 4206601 328340 0.0011 1.000 9.39 9.39 30.062 170 693 50 2083 510 1446587 4338889 354350 0.0008 1.000 8.33 8.33 34.801 180 707 64 2127 540 1508627 4524129 383200 0.0006 1.000 7.27 7.27 41.445 190 727 84 2184 570 1590878 4769856 415390 0.0005 1.000 6.55 6.55 46.406 200 750 107 2235 600 1665593 4995225 447310 0.0006 1.000 5.83 5.83 40.877 210 758 115 2375 630 1888753 5640625 499920 0.0001 1.000 5.17 5.17 85.458 220 867 224 2543 660 2168977 6466849 561060 0.0001 1.000 4.58 4.58 91.381 230 918 275 2738 690 2502622 7496644 630600 0.0002 1.000 3.99 3.99 65.952 240 953 310 2855 711 2719189 8151025 677004 0.0002 1.000 3.77 3.77 76.874 241 984 341 3.74 Flame Front Position x Time to position x t (s) time from ignition ?t ?x ?(t 2 ) ?(t) 2 ?(t.x) V f F(t) " . . e q )(. " . . tFq e 1/?V f RIFT ? Radiata Pine ? full preheat ? test 2 Ignition Time 644 m/s kW/m 2 (m/s) -0.5 10 20 30 325 40 50 60 70 80 90 100 648 4 1.000 16.73 110 650 6 1956 330 1275368 3825936 215260 0.0018 1.000 15.06 15.06 23.664 120 658 14 1973 360 1297689 3892729 236910 0.0013 1.000 13.89 13.89 27.406 130 665 21 1997 390 1329465 3988009 259770 0.0012 1.000 12.71 12.71 28.358 140 674 30 2028 420 1371222 4112784 284160 0.0008 1.000 11.59 11.59 35.000 150 689 45 2070 450 1428846 4284900 310830 0.0006 1.000 10.48 10.48 40.676 160 707 63 2120 480 1498746 4494400 339550 0.0006 1.000 9.39 9.39 41.839 170 724 80 2174 510 1576074 4726276 369940 0.0006 1.000 8.33 8.33 42.448 180 743 99 2229 540 1656869 4968441 401600 0.0005 1.000 7.27 7.27 43.589 190 762 118 2300 570 1764718 5290000 437520 0.0004 1.000 6.55 6.55 51.603 200 795 151 2376 600 1883430 5645376 475770 0.0003 1.000 5.83 5.83 53.607 210 819 175 2468 630 2032102 6091024 518870 0.0003 1.000 5.17 5.17 54.628 220 854 210 2556 660 2179766 6533136 562960 0.0003 1.000 4.58 4.58 56.651 230 883 239 2661 685 2362781 7080921 608110 0.0002 1.000 3.99 3.99 69.310 235 924 280 3.88 Flame Front Position x Time to position x t (s) time from ignition ?t ?x ?(t 2 ) ?(t) 2 ?(t.x) V f F(t) " . . e q )(. " . . tFq e 1/?V f RIFT ? Radiata Pine ? full preheat ? test 3 Ignition Time 40 m/s kW/m 2 (m/s) -0.5 10 20 30 40 50 326 60 70 80 90 100 110 649 6 15.06 120 654 11 1960 360 1280566 3841600 235280 0.0024 1.000 13.89 13.89 20.207 130 657 14 1974 390 1298934 3896676 256710 0.0021 1.000 12.71 12.71 21.602 140 663 20 1991 420 1321459 3964081 278880 0.0014 1.000 11.59 11.59 26.547 150 671 28 2015 450 1353571 4060225 302430 0.0011 1.000 10.48 10.48 30.062 160 681 38 2048 480 1398418 4194304 327930 0.0008 1.000 9.39 9.39 35.590 170 696 53 2088 510 1453698 4359744 355260 0.0007 1.000 8.33 8.33 38.730 180 711 68 2133 540 1517013 4549689 384240 0.0007 1.000 7.27 7.27 38.730 190 726 83 2186 570 1593598 4778596 415720 0.0005 1.000 6.55 6.55 43.910 200 749 106 2280 600 1736102 5198400 456790 0.0002 1.000 5.83 5.83 64.651 210 805 162 2393 630 1912947 5726449 503430 0.0002 1.000 5.17 5.17 67.747 220 839 196 2652 660 2368010 7033104 585470 0.0001 1.000 4.58 4.58 107.918 230 1008 365 2889 684 2805749 8346321 660248 0.0001 1.000 3.99 3.99 123.264 234 1042 399 3.90 Flame Front Position x Time to position x t (s) time from ignition ?t ?x ?(t 2 ) ?(t) 2 ?(t.x) V f F(t) " . . e q )(. " . . tFq e 1/?V f RIFT ? Radiata Pine ? full preheat ? test 4 Ignition Time 40 m/s kW/m 2 (m/s) -0.5 10 20 30 40 50 60 70 80 327 90 100 649 5 16.73 110 652 8 1959 330 1279269 3837681 215580 0.0021 1.000 15.06 15.06 21.602 120 658 14 1970 360 1293668 3880900 236480 0.0023 1.000 13.89 13.89 20.817 130 660 16 1988 390 1317464 3952144 258560 0.0015 1.000 12.71 12.71 26.247 140 670 26 2010 420 1346900 4040100 281600 0.0010 1.000 11.59 11.59 31.623 150 680 36 2042 450 1390164 4169764 306520 0.0009 1.000 10.48 10.48 33.212 160 692 48 2088 480 1453920 4359744 334440 0.0005 1.000 9.39 9.39 43.205 170 716 72 2145 510 1534689 4601025 365100 0.0004 1.000 8.33 8.33 47.469 180 737 93 2220 540 1644114 4928400 400110 0.0004 1.000 7.27 7.27 50.759 190 767 123 2305 570 1773059 5313025 438590 0.0003 1.000 6.55 6.55 56.605 200 801 157 2399 600 1920451 5755201 480440 0.0003 1.000 5.83 5.83 56.605 210 831 187 2482 630 2054662 6160324 521710 0.0004 1.000 5.17 5.17 49.911 220 850 206 2554 660 2175190 6522916 562300 0.0005 1.000 4.58 4.58 45.895 230 873 229 2620 690 2289238 6864400 603070 0.0004 1.000 3.99 3.99 48.480 240 897 253 2697 717 2426067 7273809 645039 0.0003 1.000 3.77 3.77 56.662 247 927 283 1.000 3.61 328 Macrocarpa Flame Front Position x Time to position x t (s) time from ignition ?t ?x ?(t 2 ) ?(t) 2 ?(t.x) V f F(t) " . . e q )(. " . . tFq e 1/?V f LIFT ? Macrocarpa Test 1 Ignition Time 291 m/s kW/m 2 (m/s) -0.5 50 75 100 125 150 306 15 1.000 21.80 21.80 175 309 18 926 525 285838 857476 162175 0.0099 1.000 20.60 20.60 10.066 200 311 20 935 600 291427 874225 187150 0.0080 1.000 18.95 18.95 11.155 225 315 24 945 675 297707 893025 212825 0.0063 1.000 17.29 17.29 12.649 250 319 28 957 750 305315 915849 239450 0.0063 1.000 15.33 15.33 12.649 275 323 32 969 825 313019 938961 266675 0.0063 1.000 13.37 13.37 12.649 300 327 36 988 900 325502 976144 296775 0.0031 1.000 11.56 11.56 17.938 325 338 47 1013 975 342277 1026169 329750 0.0024 1.000 9.75 9.75 20.502 350 348 57 1060 1050 375224 1123600 371900 0.0013 1.000 8.52 8.52 27.702 375 374 83 1105 1125 407669 1221025 415250 0.0013 1.000 7.28 7.28 27.478 400 383 92 1151 1200 441801 1324801 460900 0.0025 1.000 6.28 6.28 20.033 425 394 103 1198 1275 479166 1435204 510100 0.0012 1.000 5.27 5.27 28.371 450 421 130 1302 1350 569646 1695204 588225 0.0005 1.000 4.50 4.50 44.374 475 487 196 1471 1425 731379 2163841 702275 0.0004 1.000 3.72 3.72 53.336 500 563 272 1657 1500 922587 2745649 831500 0.0004 1.000 3.18 3.18 49.567 525 607 316 1928 1575 1259982 3717184 1017075 0.0002 1.000 2.63 2.63 65.509 550 758 467 2155 1635 1567113 4644025 1177975 0.0002 1.000 2.18 2.18 73.882 560 790 499 1.000 1.96 1.96 329 Flame Front Position x Time to position x t (s) time from ignition ?t ?x ?(t 2 ) ?(t) 2 ?(t.x) V f F(t) " . . e q )(. " . . tFq e 1/?V f LIFT ? Macrocarpa Test 2 Ignition Time 280 m/s kW/m 2 (m/s) -0.5 50 25.20 75 24.51 100 23.75 125 22.99 150 21.80 175 20.60 200 289 9 1.000 18.95 225 292 12 876 675 255810 767376 197250 0.0083 1.000 17.29 17.29 10.954 250 295 15 884 750 260498 781456 221125 0.0099 1.000 15.33 15.33 10.066 275 297 17 892 825 265234 795664 245425 0.0099 1.000 13.37 13.37 10.066 300 300 20 903 900 271845 815409 271125 0.0054 1.000 11.56 11.56 13.663 325 306 26 917 975 280357 840889 298300 0.0045 1.000 9.75 9.75 14.853 350 311 31 937 1050 292757 877969 328300 0.0035 1.000 8.52 8.52 16.959 375 320 40 963 1125 309345 927369 361650 0.0024 1.000 7.28 7.28 20.563 400 332 52 1002 1200 335124 1004004 401550 0.0016 1.000 6.28 6.28 24.658 425 350 70 1039 1275 360173 1079521 442200 0.0019 1.000 5.27 5.27 23.071 450 357 77 1086 1350 393590 1179396 489425 0.0016 1.000 4.50 4.50 25.134 475 379 99 1147 1425 440011 1315609 546175 0.0009 1.000 3.72 3.72 33.051 500 411 131 1242 1500 516866 1542564 622825 0.0007 1.000 3.18 3.18 38.307 525 452 172 1410 1575 672434 1988100 743650 0.0003 1.000 2.63 2.63 53.506 550 547 267 1.000 2.18 2.18 330 Flame Front Position x Time to position x t (s) time from ignition ?t ?x ?(t 2 ) ?(t) 2 ?(t.x) V f F(t) " . . e q )(. " . . tFq e 1/?V f LIFT ? Macrocarpa Test 3 Ignition Time 289 m/s kW/m 2 (m/s) -0.5 50 25.20 75 24.51 100 23.75 125 22.99 150 21.80 175 20.60 200 18.95 225 17.29 250 300 11 1.000 15.33 275 303 14 909 825 275445 826281 250125 0.0083 1.000 13.37 13.37 10.954 300 306 17 919 900 281545 844561 275875 0.0071 1.000 11.56 11.56 11.872 325 310 21 929 975 287705 863041 302100 0.0071 1.000 9.75 9.75 11.872 350 313 24 946 1050 298398 894916 331425 0.0035 1.000 8.52 8.52 16.886 375 323 34 969 1125 313187 938961 363875 0.0025 1.000 7.28 7.28 20.000 400 333 44 1002 1200 334934 1004004 401375 0.0022 1.000 6.28 6.28 21.508 425 346 57 1047 1275 366029 1096209 445850 0.0014 1.000 5.27 5.27 26.747 450 368 79 1107 1350 409589 1225449 499325 0.0011 1.000 4.50 4.50 30.680 475 393 104 1214 1425 495082 1473796 578775 0.0006 1.000 3.72 3.72 42.380 500 453 164 1330 1500 593914 1768900 667275 0.0005 1.000 3.18 3.18 43.378 525 484 195 1451 1562 703661 2105401 756618 0.0006 1.000 2.63 2.63 40.566 537 514 225 1.000 2.42 2.42 Flame Front Position x Time to position x t (s) time from ignition ?t ?x ?(t 2 ) ?(t) 2 ?(t.x) V f F(t) " . . e q )(. " . . tFq e 1/?V f RIFT ? Macrocarpa Low preheat ? ISO ign- Test 1 Ignition Time 63 m/s kW/m 2 (m/s) -0.5 10 50 0 33.88 331 20 51 1 153 60 7805 23409 3080 0.0100 0.342 32.87 11.25 10.000 30 52 2 156 90 8114 24336 4700 0.0100 0.346 32.16 11.11 10.000 40 53 3 159 120 8429 25281 6380 0.0100 0.349 29.67 10.35 10.000 50 54 4 162 150 8750 26244 8120 0.0100 0.352 27.17 9.57 10.000 60 55 5 165 180 9077 27225 9920 0.0100 0.355 24.95 8.87 10.000 70 56 6 168 210 9410 28224 11780 0.0100 0.359 22.99 8.24 10.000 80 57 7 175 240 10229 30625 14060 0.0029 0.362 21.03 7.61 18.559 90 62 12 184 270 11318 33856 16640 0.0024 0.377 18.88 7.12 20.207 100 65 15 196 300 12830 38416 19670 0.0028 0.386 16.73 6.46 18.772 110 69 19 211 330 14915 44521 23330 0.0016 0.398 15.06 5.99 24.944 120 77 27 238 360 19154 56644 28790 0.0008 0.420 13.89 5.84 34.431 130 92 42 276 390 25842 76176 36180 0.0007 0.460 12.71 5.84 38.730 140 107 57 343 420 40649 117649 48540 0.0004 0.496 11.59 5.75 52.489 150 144 94 413 450 58429 170569 62500 0.0003 0.575 10.48 6.02 53.473 160 162 112 479 480 76909 229441 76930 0.0007 0.610 9.39 5.73 38.447 170 173 123 520 510 90398 270400 88630 0.0009 0.630 8.33 5.25 33.922 180 185 135 630 540 138138 396900 114390 0.0002 0.652 7.27 4.74 76.792 190 272 222 754 570 196418 568516 144380 0.0002 0.790 6.55 5.18 78.562 200 297 247 889 600 264593 790321 178280 0.0004 0.826 5.83 4.81 49.004 210 320 270 959 630 307573 919681 201840 0.0004 0.857 5.17 4.43 47.438 220 342 292 1023 651 349685 1046529 222221 0.0003 0.886 4.58 4.06 60.505 221 361 311 0.910 4.52 4.12 Flame Front Position x Time to position x t (s) time from ignition ?t ?x ?(t 2 ) ?(t) 2 ?(t.x) V f F(t) " . . e q )(. " . . tFq e 1/?V f RIFT ? Macrocarpa Low preheat ? ISO ign- Test 2 Ignition Time 50 m/s kW/m 2 (m/s) -0.5 10 63 0 33.88 20 64 1 193 60 12421 37249 3890 0.0064 0.383 32.87 12.60 12.472 30 66 3 198 90 13076 39204 5980 0.0050 0.389 32.16 12.52 14.142 40 68 5 203 120 13741 41209 8150 0.0064 0.395 29.67 11.72 12.472 50 69 6 210 150 14714 44100 10550 0.0036 0.398 27.17 10.82 16.733 332 60 73 10 221 180 16331 48841 13360 0.0020 0.409 24.95 10.21 22.509 70 79 16 240 210 19314 57600 16950 0.0013 0.426 22.99 9.79 27.568 80 88 25 267 240 23985 71289 21570 0.0009 0.449 21.03 9.45 32.514 90 100 37 302 270 30740 91204 27440 0.0008 0.479 18.88 9.05 36.091 100 114 51 349 300 41221 121801 35250 0.0006 0.512 16.73 8.56 42.111 110 135 72 406 330 55870 164836 45090 0.0005 0.557 15.06 8.38 46.372 120 157 94 479 360 77843 229441 58000 0.0004 0.600 13.89 8.34 51.191 130 187 124 568 390 109794 322624 74510 0.0003 0.655 12.71 8.33 57.984 140 224 161 673 420 153789 452929 94970 0.0003 0.717 11.59 8.31 61.239 150 262 199 808 450 222504 652864 122180 0.0002 0.776 10.48 8.12 70.586 160 322 259 948 480 304824 898704 152700 0.0002 0.860 9.39 8.07 71.784 170 364 301 1105 510 411741 1221025 188820 0.0002 0.914 8.33 7.62 69.850 180 419 356 1307 535 582633 1708249 234240 0.0001 0.981 7.27 7.13 106.818 185 524 461 1.000 6.91 6.91 Flame Front Position x Time to position x t (s) time from ignition ?t ?x ?(t 2 ) ?(t) 2 ?(t.x) V f F(t) " . . e q )(. " . . tFq e 1/?V f RIFT ? Macrocarpa Low preheat ? ISO ign- Test 3 Ignition Time 40 m/s kW/m 2 (m/s) -0.5 10 40 0 33.88 20 41 1 124 60 5130 15376 2510 0.0064 0.307 32.87 10.09 12.472 30 43 3 128 90 5466 16384 3870 0.0064 0.314 32.16 10.10 12.472 40 44 4 132 120 5810 17424 5300 0.0100 0.318 29.67 9.43 10.000 50 45 5 135 150 6077 18225 6770 0.0100 0.321 27.17 8.73 10.000 60 46 6 141 180 6641 19881 8510 0.0036 0.325 24.95 8.11 16.733 70 50 10 152 210 7752 23104 10740 0.0020 0.339 22.99 7.79 22.509 80 56 16 168 240 9480 28224 13560 0.0017 0.359 21.03 7.54 24.495 90 62 22 188 270 11880 35344 17060 0.0014 0.377 18.88 7.12 26.547 100 70 30 217 300 15969 47089 21930 0.0008 0.401 16.73 6.71 34.431 110 85 45 254 330 21926 64516 28230 0.0007 0.442 15.06 6.65 38.086 120 99 59 305 360 31667 93025 36960 0.0005 0.477 13.89 6.62 42.774 130 121 81 358 390 43486 128164 46930 0.0005 0.527 12.71 6.70 44.280 333 140 138 98 423 420 60581 178929 59650 0.0005 0.563 11.59 6.53 46.705 150 164 124 501 450 85541 251001 75760 0.0003 0.614 10.48 6.43 55.427 160 199 159 570 480 109346 324900 91630 0.0004 0.676 9.39 6.35 49.321 170 207 167 664 510 149014 440896 113470 0.0003 0.689 8.33 5.74 58.926 180 258 218 751 540 191209 564001 135970 0.0002 0.770 7.27 5.60 63.731 190 286 246 871 570 255289 758641 166180 0.0003 0.810 6.55 5.31 59.083 200 327 287 965 594 312629 931225 191548 0.0002 0.866 5.83 5.05 68.160 204 352 312 0.899 5.57 5.00 Flame Front Position x Time to position x t (s) time from ignition ?t ?x ?(t 2 ) ?(t) 2 ?(t.x) V f F(t) " . . e q )(. " . . tFq e 1/?V f RIFT ? Macrocarpa Low preheat ? ISO ign- Test 4 Ignition Time 40 m/s kW/m 2 (m/s) -0.5 10 40 0 33.88 20 41 1 123 60 5045 15129 2480 0.0100 0.307 32.87 10.09 10.000 30 42 2 128 90 5470 16384 3880 0.0046 0.311 32.16 9.99 14.720 40 45 5 134 120 5998 17956 5410 0.0039 0.321 29.67 9.54 15.916 50 47 7 145 150 7043 21025 7330 0.0023 0.328 27.17 8.93 20.817 60 53 13 155 180 8043 24025 9380 0.0023 0.349 24.95 8.70 20.817 70 55 15 166 210 9198 27556 11670 0.0039 0.355 22.99 8.17 15.916 80 58 18 185 240 11573 34225 14970 0.0010 0.365 21.03 7.68 31.123 90 72 32 211 270 15109 44521 19220 0.0009 0.407 18.88 7.68 34.178 100 81 41 243 300 19845 59049 24480 0.0011 0.431 16.73 7.21 30.000 110 90 50 301 330 31561 90601 33600 0.0004 0.455 15.06 6.85 52.696 120 130 90 370 360 47500 136900 45000 0.0003 0.546 13.89 7.59 55.777 130 150 110 456 390 70376 207936 59740 0.0004 0.587 12.71 7.46 48.094 140 176 136 537 420 97997 288369 75790 0.0003 0.636 11.59 7.37 55.427 150 211 171 627 450 133097 393129 94690 0.0003 0.696 10.48 7.29 56.651 160 240 200 724 480 176650 524176 116460 0.0003 0.742 9.39 6.97 55.716 170 273 233 853 510 247729 727609 146010 0.0002 0.792 8.33 6.60 72.060 180 340 300 992 540 333770 984064 179620 0.0002 0.883 7.27 6.43 73.643 334 190 379 339 0.933 6.55 Flame Front Position x Time to position x t (s) time from ignition ?t ?x ?(t 2 ) ?(t) 2 ?(t.x) V f F(t) " . . e q )(. " . . tFq e 1/?V f RIFT ? Macrocarpa Low preheat ? ISO ign- Test 5 Ignition Time 40 m/s kW/m 2 (m/s) -0.5 10 40 0 33.88 20 43 3 129 60 5565 16641 2640 0.0033 0.314 32.87 10.33 17.321 30 46 6 136 90 6174 18496 4120 0.0046 0.325 32.16 10.45 14.720 40 47 7 142 120 6726 20164 5710 0.0064 0.328 29.67 9.74 12.472 50 49 9 148 150 7314 21904 7450 0.0039 0.335 27.17 9.11 15.916 60 52 12 156 180 8130 24336 9420 0.0033 0.346 24.95 8.62 17.321 70 55 15 169 210 9573 28561 11930 0.0019 0.355 22.99 8.17 22.949 80 62 22 187 240 11769 34969 15110 0.0013 0.377 21.03 7.94 27.406 90 70 30 213 270 15305 45369 19360 0.0010 0.401 18.88 7.57 30.950 100 81 41 246 300 20486 60516 24850 0.0008 0.431 16.73 7.21 35.440 110 95 55 284 330 27250 80656 31510 0.0007 0.467 15.06 7.03 36.751 120 108 68 320 360 34378 102400 38620 0.0009 0.498 13.89 6.91 33.348 130 117 77 355 390 42253 126025 46370 0.0009 0.518 12.71 6.59 33.348 140 130 90 392 420 51614 153664 55160 0.0007 0.546 11.59 6.33 37.448 150 145 105 438 450 64494 191844 66030 0.0006 0.577 10.48 6.04 40.676 160 163 123 502 480 85230 252004 80810 0.0004 0.612 9.39 5.74 50.075 170 194 154 587 510 117105 344569 100460 0.0003 0.667 8.33 5.56 57.933 180 230 190 681 533 156585 463761 121411 0.0002 0.727 7.27 5.29 68.972 183 257 217 0.768 7.06 Flame Front Position x Time to position x t (s) time from ignition ?t ?x ?(t 2 ) ?(t) 2 ?(t.x) V f F(t) " . . e q )(. " . . tFq e 1/?V f RIFT ? Macrocarpa Low preheat ? ISO ign- Test 6 Ignition Time 40 m/s kW/m 2 (m/s) -0.5 10 40 0 33.88 335 20 42 2 126 60 5300 15876 2560 0.0050 0.311 32.87 10.21 14.142 30 44 4 131 90 5725 17161 3960 0.0064 0.318 32.16 10.22 12.472 40 45 5 137 120 6265 18769 5520 0.0046 0.321 29.67 9.54 14.720 50 48 8 143 150 6829 20449 7200 0.0039 0.332 27.17 9.02 15.916 60 50 10 154 180 7940 23716 9320 0.0023 0.339 24.95 8.45 20.817 70 56 16 170 210 9732 28900 12040 0.0014 0.359 22.99 8.24 26.547 80 64 24 198 240 13316 39204 16060 0.0009 0.383 21.03 8.06 33.575 90 78 38 240 270 19784 57600 21940 0.0006 0.423 18.88 7.99 41.445 100 98 58 295 300 29849 87025 29910 0.0005 0.474 16.73 7.93 45.281 110 119 79 364 330 45374 132496 40530 0.0004 0.523 15.06 7.87 49.666 120 147 107 437 360 65011 190969 52960 0.0004 0.581 13.89 8.07 51.040 130 171 131 522 390 92466 272484 68430 0.0003 0.627 12.71 7.96 53.607 140 204 164 611 420 126553 373321 86190 0.0003 0.684 11.59 7.93 57.011 150 236 196 716 450 173488 512656 108120 0.0003 0.736 10.48 7.71 60.123 160 276 236 852 480 247472 725904 137360 0.0002 0.796 9.39 7.47 72.748 170 340 300 994 510 334660 988036 170000 0.0002 0.883 8.33 7.36 72.183 180 378 338 1138 540 434884 1295044 205640 0.0002 0.932 7.27 6.78 63.272 190 420 380 1261 565 533653 1590121 238125 0.0002 0.982 6.55 6.43 75.328 195 463 423 6.19 Flame Front Position x Time to position x t (s) time from ignition ?t ?x ?(t 2 ) ?(t) 2 ?(t.x) V f F(t) " . . e q )(. " . . tFq e 1/?V f RIFT ? Macrocarpa ? full preheat ? test 1 Ignition Time 436 m/s kW/m 2 (m/s) -0.5 10 20 30 40 50 60 70 441 5 1.000 22.99 336 80 446 10 1334 240 593206 1779556 106780 0.0029 1.000 21.03 21.03 18.559 90 447 11 1342 270 600326 1800964 120810 0.0064 1.000 18.88 18.88 12.472 100 449 13 1350 300 607526 1822500 135070 0.0027 1.000 16.73 16.73 19.272 110 454 18 1363 330 619317 1857769 150040 0.0018 1.000 15.06 15.06 23.484 120 460 24 1384 360 638616 1915456 166240 0.0012 1.000 13.89 13.89 28.577 130 470 34 1411 390 663861 1990921 183640 0.0010 1.000 12.71 12.71 32.416 140 481 45 1446 420 697286 2090916 202690 0.0008 1.000 11.59 11.59 35.440 150 495 59 1504 450 755170 2262016 226070 0.0004 1.000 10.48 10.48 49.780 160 528 92 1570 480 823018 2464900 251720 0.0004 1.000 9.39 9.39 51.603 170 547 111 1645 510 902893 2706025 280070 0.0005 1.000 8.33 8.33 45.895 180 570 134 1712 540 978134 2930944 308640 0.0004 1.000 7.27 7.27 49.004 190 595 159 1792 570 1072054 3211264 341050 0.0003 1.000 6.55 6.55 53.519 200 627 191 1882 594 1182754 3541924 373090 0.0002 1.000 5.83 5.83 68.216 204 660 224 5.57 Flame Front Position x Time to position x t (s) time from ignition ?t ?x ?(t 2 ) ?(t) 2 ?(t.x) V f F(t) " . . e q )(. " . . tFq e 1/?V f RIFT ? Macrocarpa ? full preheat ? test 2 Ignition Time 438 m/s kW/m 2 (m/s) -0.5 10 20 30 40 50 60 70 445 7 1.000 22.99 80 448 10 1344 240 602130 1806336 107580 0.0033 1.000 21.03 21.03 17.321 90 451 13 1352 270 609314 1827904 121730 0.0039 1.000 18.88 18.88 15.916 100 453 15 1360 300 616546 1849600 136050 0.0039 1.000 16.73 16.73 15.916 110 456 18 1369 330 624745 1874161 150660 0.0028 1.000 15.06 15.06 18.772 120 460 22 1381 360 635761 1907161 165810 0.0022 1.000 13.89 13.89 21.257 130 465 27 1402 390 655354 1965604 182430 0.0011 1.000 12.71 12.71 29.967 337 140 477 39 1435 420 686803 2059225 201180 0.0007 1.000 11.59 11.59 37.544 150 493 55 1473 450 723587 2169729 221210 0.0008 1.000 10.48 10.48 36.374 160 503 65 1510 480 760254 2280100 241810 0.0010 1.000 9.39 9.39 32.416 170 514 76 1551 510 802361 2405601 263980 0.0006 1.000 8.33 8.33 39.919 180 534 96 1613 540 868577 2601769 290850 0.0004 1.000 7.27 7.27 50.888 190 565 127 1725 570 996257 2975625 328670 0.0002 1.000 6.55 6.55 69.015 200 626 188 1840 600 1132302 3385600 368840 0.0002 1.000 5.83 5.83 66.981 210 649 211 1951 630 1270053 3806401 410210 0.0004 1.000 5.17 5.17 50.053 220 676 238 1.000 4.58 Flame Front Position x Time to position x t (s) time from ignition ?t ?x ?(t 2 ) ?(t) 2 ?(t.x) V f F(t) " . . e q )(. " . . tFq e 1/?V f RIFT ? Macrocarpa ? full preheat ? test 3 Ignition Time 438 m/s kW/m 2 (m/s) -0.5 10 20 30 40 50 60 70 446 8 1.000 22.99 80 447 9 1341 240 599429 1798281 107300 0.0100 1.000 21.03 21.03 10.000 90 448 10 1344 270 602114 1806336 120980 0.0100 1.000 18.88 18.88 10.000 100 449 11 1347 300 604805 1814409 134720 0.0100 1.000 16.73 16.73 10.000 110 450 12 1351 330 608405 1825201 148640 0.0064 1.000 15.06 15.06 12.472 120 452 14 1356 360 612920 1838736 162760 0.0050 1.000 13.89 13.89 14.142 130 454 16 1362 390 618356 1855044 177100 0.0050 1.000 12.71 12.71 14.142 140 456 18 1369 420 624733 1874161 191710 0.0039 1.000 11.59 11.59 15.916 150 459 21 1381 450 635773 1907161 207250 0.0019 1.000 10.48 10.48 22.949 160 466 28 1402 480 655366 1965604 224500 0.0011 1.000 9.39 9.39 30.246 170 477 39 1438 510 689710 2067844 244750 0.0007 1.000 8.33 8.33 38.447 180 495 57 1516 540 768490 2298256 273550 0.0003 1.000 7.27 7.27 59.909 338 190 544 106 1601 570 856805 2563201 304860 0.0003 1.000 6.55 6.55 59.909 200 562 124 5.83 Flame Front Position x Time to position x t (s) time from ignition ?t ?x ?(t 2 ) ?(t) 2 ?(t.x) V f F(t) " . . e q )(. " . . tFq e 1/?V f RIFT ? Macrocarpa ? full preheat ? test 4 Ignition Time 443 m/s kW/m 2 (m/s) -0.5 10 20 30 40 50 60 70 448 5 22.99 80 450 7 1351 240 608413 1825201 108130 0.0039 1.000 21.03 21.03 15.916 90 453 10 1359 270 615645 1846881 122370 0.0033 1.000 18.88 18.88 17.321 100 456 13 1368 300 623826 1871424 136860 0.0033 1.000 16.73 16.73 17.321 110 459 16 1375 330 630217 1890625 151290 0.0046 1.000 15.06 15.06 14.720 120 460 17 1381 360 635725 1907161 165750 0.0064 1.000 13.89 13.89 12.472 130 462 19 1387 390 641269 1923769 180360 0.0039 1.000 12.71 12.71 15.916 140 465 22 1397 420 650569 1951609 195660 0.0024 1.000 11.59 11.59 20.207 150 470 27 1409 450 661801 1985281 211440 0.0022 1.000 10.48 10.48 21.257 160 474 31 1423 480 675017 2024929 227770 0.0022 1.000 9.39 9.39 21.257 170 479 36 1446 510 697166 2090916 246010 0.0010 1.000 8.33 8.33 31.954 180 493 50 1488 540 738746 2214144 268210 0.0005 1.000 7.27 7.27 43.434 190 516 73 1543 570 794461 2380849 293580 0.0005 1.000 6.55 6.55 45.389 200 534 91 1602 600 856116 2566404 320760 0.0006 1.000 5.83 5.83 42.426 210 552 109 1656 630 914760 2742336 348120 0.0006 1.000 5.17 5.17 42.426 220 570 127 1719 657 986013 2954961 376839 0.0004 1.000 4.58 4.58 52.099 227 597 154 4.17 339 Rimu Flame Front Position x Time to position x t (s) time from ignition ?t ?x ?(t 2 ) ?(t) 2 ?(t.x) V f F(t) " . . e q )(. " . . tFq e 1/?V f LIFT Rimu test 1 Ignition Time 715 m/s kW/m 2 (m/s) -0.5 50 #N/A 75 26.45 100 720 5 1.000 26.10 26.10 125 721 6 2165 375 1562417 4687225 270725 0.0115 1.000 25.30 25.30 9.309 150 724 9 2171 450 1571093 4713241 325775 0.0099 1.000 24.50 24.50 10.066 175 726 11 2178 525 1581236 4743684 381250 0.0125 1.000 23.20 23.20 8.944 200 728 13 2191 600 1600229 4800481 438475 0.0040 1.000 21.90 21.90 15.802 225 737 22 2206 675 1622234 4866436 496675 0.0037 1.000 20.70 20.70 16.517 250 741 26 2223 750 1647275 4941729 555950 0.0063 1.000 19.50 19.50 12.649 275 745 30 2237 825 1668107 5004169 615425 0.0049 1.000 18.00 18.00 14.236 300 751 36 2251 900 1689051 5067001 675550 0.0049 1.000 16.50 16.50 14.236 325 755 40 2269 975 1716195 5148361 737725 0.0040 1.000 14.85 14.85 15.776 350 763 48 2300 1050 1763718 5290000 805675 0.0018 1.000 13.20 13.20 23.872 375 782 67 2367 1117 1869377 5602689 882524 0.0007 1.000 11.50 11.50 38.703 392 822 107 1.000 10.34 10.34 Flame Front Position x Time to position x t (s) time from ignition ?t ?x ?(t 2 ) ?(t) 2 ?(t.x) V f F(t) " . . e q )(. " . . tFq e 1/?V f LIFT Rimu test 2 Ignition Time 717 m/s kW/m 2 (m/s) -0.5 50 75 100 125 150 175 340 200 729 12 1.000 21.90 21.90 225 731 14 2193 675 1603091 4809249 493525 0.0125 1.000 20.70 20.70 8.944 250 733 16 2201 750 1614819 4844401 550400 0.0080 1.000 19.50 19.50 11.155 275 737 20 2213 825 1632507 4897369 608825 0.0049 1.000 18.00 18.00 14.236 300 743 26 2227 900 1653227 4959529 668350 0.0049 1.000 16.50 16.50 14.236 325 747 30 2244 975 1678574 5035536 729575 0.0044 1.000 14.85 14.85 15.015 350 754 37 2269 1050 1716349 5148361 794675 0.0023 1.000 13.20 13.20 20.870 375 768 51 2308 1125 1776136 5326864 866300 0.0016 1.000 11.50 11.50 25.364 400 786 69 2370 1200 1873476 5616900 949200 0.0010 1.000 9.80 9.80 31.305 425 816 99 2458 1275 2016388 6041764 1046400 0.0007 1.000 8.35 8.35 37.544 450 856 139 2576 1350 2215808 6635776 1161400 0.0006 1.000 6.90 6.90 42.010 475 904 187 2752 1425 2534016 7573504 1310600 0.0004 1.000 5.95 5.95 52.900 500 992 275 1.000 5.00 5.00 Flame Front Position x Time to position x t (s) time from ignition ?t ?x ?(t 2 ) ?(t) 2 ?(t.x) V f F(t) " . . e q )(. " . . tFq e 1/?V f LIFT Rimu test 3 Ignition Time 736 m/s kW/m 2 (m/s) -0.5 50 75 100 125 150 175 200 225 250 275 300 750 14 1.000 16.50 16.50 325 761 25 2282 975 1736062 5207524 742175 0.0024 1.000 14.85 14.85 20.502 350 771 35 2325 1050 1802411 5405625 814550 0.0015 1.000 13.20 13.20 25.884 375 793 57 2387 1125 1900619 5697769 896425 0.0010 1.000 11.50 11.50 32.376 341 400 823 87 2492 1200 2073554 6210064 998875 0.0006 1.000 9.80 9.80 41.261 425 876 140 2639 1265 2328305 6964321 1115100 0.0003 1.000 8.35 8.35 54.376 440 940 204 1.000 7.48 7.48 Flame Front Position x Time to position x t (s) time from ignition ?t ?x ?(t 2 ) ?(t) 2 ?(t.x) V f F(t) " . . e q )(. " . . tFq e 1/?V f RIFT ? Rimu ? Low preheat ? ISO ign data ? Test 1 Ignition Time 80 m/s kW/m 2 (m/s) -0.5 10 60 0 33.88 20 61 1 184 60 11290 33856 3710 0.0064 0.392 32.87 12.89 12.472 30 63 3 189 90 11915 35721 5710 0.0050 0.399 32.16 12.82 14.142 40 65 5 197 120 12955 38809 7940 0.0032 0.405 29.67 12.01 17.638 50 69 9 208 150 14462 43264 10490 0.0022 0.417 27.17 11.33 21.257 60 74 14 227 180 17293 51529 13770 0.0013 0.432 24.95 10.78 27.889 70 84 24 249 210 20813 62001 17600 0.0012 0.460 22.99 10.58 29.306 80 91 31 276 240 25538 76176 22250 0.0012 0.479 21.03 10.07 29.306 90 101 41 302 270 30582 91204 27370 0.0011 0.505 18.88 9.53 30.836 100 110 50 330 300 36462 108900 33180 0.0011 0.527 16.73 8.81 30.000 110 119 59 358 330 42902 128164 39570 0.0011 0.548 15.06 8.25 30.836 120 129 69 393 360 51827 154449 47420 0.0008 0.570 13.89 7.92 36.374 130 145 85 434 390 63266 188356 56730 0.0006 0.605 12.71 7.68 39.377 140 160 100 492 415 81594 242064 68365 0.0003 0.635 11.59 7.36 54.502 145 187 127 0.687 11.59 7.96 342 Flame Front Position x Time to position x t (s) time from ignition ?t ?x ?(t 2 ) ?(t) 2 ?(t.x) V f F(t) " . . e q )(. " . . tFq e 1/?V f RIFT ? Rimu ? Low preheat ? ISO ign data ? Test 2 Ignition Time 80 m/s kW/m 2 (m/s) -0.5 10 63 0 33.88 20 64 1 193 60 12421 37249 3890 0.0064 0.402 32.87 13.20 12.472 30 66 3 199 90 13213 39601 6020 0.0039 0.408 32.16 13.12 15.916 40 69 6 207 120 14301 42849 8340 0.0033 0.417 29.67 12.37 17.321 50 72 9 218 150 15874 47524 10980 0.0024 0.426 27.17 11.58 20.207 60 77 14 232 180 18002 53824 14030 0.0018 0.441 24.95 10.99 23.484 70 83 20 253 210 21467 64009 17870 0.0012 0.457 22.99 10.52 28.577 80 93 30 281 240 26563 78961 22700 0.0009 0.484 21.03 10.18 33.212 90 105 42 316 270 33598 99856 28690 0.0008 0.514 18.88 9.71 35.365 100 118 55 356 300 42638 126736 35880 0.0007 0.545 16.73 9.12 37.448 110 133 70 402 330 54414 161604 44550 0.0006 0.579 15.06 8.72 40.676 120 151 88 449 360 67715 201601 54200 0.0006 0.617 13.89 8.57 40.104 130 165 102 505 389 85747 255025 65841 0.0005 0.645 12.71 8.20 45.339 139 189 126 0.690 12.71 8.77 Flame Front Position x Time to position x t (s) time from ignition ?t ?x ?(t 2 ) ?(t) 2 ?(t.x) V f F(t) " . . e q )(. " . . tFq e 1/?V f RIFT ? Rimu ? Low preheat ? ISO ign data ? Test 3 Ignition Time 69 m/s kW/m 2 (m/s) -0.5 10 63 0 33.88 20 64 1 192 60 12290 36864 3860 0.0100 0.402 32.87 13.20 10.000 30 65 2 199 90 13221 39601 6030 0.0029 0.405 32.16 13.02 18.559 40 70 7 212 120 15054 44944 8600 0.0017 0.420 29.67 12.46 24.608 50 77 14 230 150 17718 52900 11630 0.0015 0.441 27.17 11.97 25.520 60 83 20 246 180 20214 60516 14850 0.0021 0.457 24.95 11.41 21.602 70 86 23 259 210 22385 67081 18200 0.0028 0.466 22.99 10.71 18.772 343 80 90 27 277 240 25697 76729 22310 0.0012 0.476 21.03 10.02 28.363 90 101 38 312 270 32942 97344 28390 0.0006 0.505 18.88 9.53 39.919 100 121 58 360 300 43886 129600 36370 0.0005 0.552 16.73 9.24 43.059 110 138 75 411 330 56789 168921 45520 0.0006 0.590 15.06 8.88 39.431 120 152 89 465 355 72773 216225 55295 0.0004 0.619 13.89 8.59 50.845 125 175 112 0.664 13.89 9.22 Flame Front Position x Time to position x t (s) time from ignition ?t ?x ?(t 2 ) ?(t) 2 ?(t.x) V f F(t) " . . e q )(. " . . tFq e 1/?V f RIFT ? Rimu ? Low preheat ? ISO ign data ? Test 4 Ignition Time 66 m/s kW/m 2 (m/s) -0.5 10 60 0 33.88 20 61 1 183 60 11165 33489 3680 0.0100 0.392 32.87 12.89 10.000 30 62 2 187 90 11661 34969 5640 0.0064 0.395 32.16 12.71 12.472 40 64 4 191 120 12165 36481 7670 0.0064 0.402 29.67 11.92 12.472 50 65 5 195 150 12677 38025 9770 0.0100 0.405 27.17 11.00 10.000 60 66 6 201 180 13481 40401 12110 0.0036 0.408 24.95 10.18 16.733 70 70 10 210 210 14732 44100 14780 0.0025 0.420 22.99 9.66 20.000 80 74 14 224 240 16776 50176 18020 0.0020 0.432 21.03 9.08 22.509 90 80 20 242 270 19620 58564 21920 0.0014 0.449 18.88 8.48 26.547 100 88 28 265 300 23553 70225 26670 0.0012 0.471 16.73 7.88 29.172 110 97 37 291 330 28389 84681 32190 0.0011 0.494 15.06 7.45 30.000 120 106 46 320 360 34334 102400 38600 0.0010 0.517 13.89 7.18 31.675 130 117 57 354 390 42086 125316 46270 0.0008 0.543 12.71 6.90 35.440 140 131 71 391 420 51299 152881 55000 0.0008 0.575 11.59 6.66 36.091 150 143 83 431 450 62259 185761 64910 0.0008 0.600 10.48 6.29 36.091 160 157 97 468 480 73322 219024 75130 0.0008 0.629 9.39 5.91 35.440 170 168 108 522 505 91682 272484 88155 0.0003 0.651 8.33 5.42 54.740 175 197 137 0.705 7.80 5.50 344 Flame Front Position x Time to position x t (s) time from ignition ?t ?x ?(t 2 ) ?(t) 2 ?(t.x) V f F(t) " . . e q )(. " . . tFq e 1/?V f RIFT ? Rimu ? Low preheat ? ISO ign data ? Test 5 Ignition Time 67 m/s kW/m 2 (m/s) -0.5 10 50 0 33.88 20 51 1 153 60 7805 23409 3080 0.0100 0.359 32.87 11.79 10.000 30 52 2 156 90 8114 24336 4700 0.0100 0.362 32.16 11.64 10.000 40 53 3 160 120 8538 25600 6430 0.0064 0.366 29.67 10.84 12.472 50 55 5 173 150 10059 29929 8770 0.0015 0.372 27.17 10.12 26.247 60 65 15 194 180 12726 37636 11830 0.0011 0.405 24.95 10.10 30.836 70 74 24 224 210 16926 50176 15880 0.0010 0.432 22.99 9.93 31.675 80 85 35 251 240 21165 63001 20260 0.0011 0.463 21.03 9.74 30.246 90 92 42 283 270 26925 80089 25680 0.0009 0.482 18.88 9.09 32.998 100 106 56 331 300 37389 109561 33510 0.0005 0.517 16.73 8.65 46.029 110 133 83 398 330 54206 158404 44310 0.0004 0.579 15.06 8.72 51.481 120 159 109 478 362 77566 228484 58262 0.0004 0.633 13.89 8.79 49.071 132 186 136 0.685 12.71 There was insufficient flame spread for a full ?preheat result for Rimu in the RIFT 345 NZ Beech Flame Front Position x Time to position x t (s) time from ignition ?t ?x ?(t 2 ) ?(t) 2 ?(t.x) V f F(t) " . . e q )(. " . . tFq e 1/?V f LIFT ? Beech ? Test 1 Ignition Time 415 m/s kW/m 2 (m/s) -0.5 50 7 100 125 150 175 200 225 427 12 1.000 17.29 17.29 250 429 14 1296 750 559970 1679616 324325 0.0033 1.000 15.55 15.55 17.365 275 440 25 1319 825 580141 1739761 363250 0.0024 1.000 13.37 13.37 20.502 300 450 35 1359 900 616061 1846881 408425 0.0017 1.000 11.19 11.19 24.467 325 469 54 1415 975 668477 2002225 461025 0.0011 1.000 9.75 9.75 30.484 350 496 81 1493 1050 744761 2229049 524025 0.0008 1.000 8.32 8.32 34.392 375 528 113 1615 1125 874081 2608225 608000 0.0005 1.000 7.28 7.28 44.356 400 591 176 1786 1200 1072954 3189796 717875 0.0004 1.000 6.25 6.25 52.803 425 667 252 1976 1275 1309694 3904576 842975 0.0004 1.000 5.27 5.27 50.723 450 718 303 2206 1350 1634454 4866436 996550 0.0003 1.000 4.29 4.29 56.543 475 821 406 2457 1414 2032289 6036849 1161977 0.0002 1.000 3.72 3.72 71.521 489 918 503 1.000 3.72 346 Flame Front Position x Time to position x t (s) time from ignition ?t ?x ?(t 2 ) ?(t) 2 ?(t.x) V f F(t) " . . e q )(. " . . tFq e 1/?V f LIFT ? Beech ? Test 2 Ignition Time 397 m/s kW/m 2 (m/s) -0.5 50 7 100 416 1 834 300 347780 695556 93850 0.00009 1.000 23.82 23.82 125 418 21 1254 375 524180 1572516 156850 0.01250 1.000 22.99 22.99 8.944 150 420 23 1260 450 529208 1587600 189100 0.01250 1.000 22.16 22.16 8.944 175 422 25 1267 525 535109 1605289 221850 0.00987 1.000 20.60 20.60 10.066 200 425 28 1274 600 541038 1623076 254925 0.00987 1.000 19.04 19.04 10.066 225 427 30 1282 675 547854 1643524 288575 0.00987 1.000 17.29 17.29 10.066 250 430 33 1302 750 565254 1695204 325950 0.00242 1.000 15.55 15.55 20.331 275 445 48 1332 825 591774 1774224 366975 0.00184 1.000 13.37 13.37 23.286 300 457 60 1373 900 628715 1885129 412550 0.00192 1.000 11.19 11.19 22.826 325 471 74 1422 975 674726 2022084 463075 0.00133 1.000 9.75 9.75 27.470 350 494 97 1487 1050 738361 2211169 521725 0.00098 1.000 8.32 8.32 31.989 375 522 125 1579 1125 833489 2493241 593850 0.00072 1.000 7.28 7.28 37.367 400 563 166 1706 1200 975094 2910436 684875 0.00050 1.000 6.25 6.25 44.715 425 621 224 1870 1262 1173206 3496900 788907 0.00030 1.000 5.27 5.27 57.881 437 686 289 1.000 5.27 LIFT test 3 Flame Front Position x Time to position x t (s) time from ignition ?t ?x ?(t 2 ) ?(t) 2 ?(t.x) V f F(t) " . . e q )(. " . . tFq e 1/?V f Ignition Time 397 m/s kW/m 2 (m/s) -0.5 50 7 100 125 347 150 412 15 1.000 22.16 175 413 16 1242 525 514202 1542564 217475 0.0089 1.000 20.60 20.60 10.583 200 417 20 1250 600 520858 1562500 250175 0.0071 1.000 19.04 19.04 11.872 225 420 23 1270 675 537778 1612900 286150 0.0028 1.000 17.29 17.29 19.018 250 433 36 1294 750 558370 1674436 324025 0.0023 1.000 15.55 15.55 20.687 275 441 44 1333 825 592651 1776889 367225 0.0018 1.000 13.37 13.37 23.359 300 459 62 1382 900 637486 1909924 415625 0.0012 1.000 11.19 11.19 28.707 325 482 85 1448 975 700054 2096704 471800 0.0010 1.000 9.75 9.75 30.993 350 507 110 1524 1050 775598 2322576 534725 0.0009 1.000 8.32 8.32 32.575 375 535 138 1612 1125 868174 2598544 606075 0.0008 1.000 7.28 7.28 35.569 400 570 173 1725 1200 995525 2975625 692125 0.0006 1.000 6.25 6.25 41.445 425 620 223 1893 1275 1203509 3583449 807850 0.0004 1.000 5.27 5.27 52.102 450 703 306 1.000 4.29 Flame Front Position x Time to position x t (s) time from ignition ?t ?x ?(t 2 ) ?(t) 2 ?(t.x) V f F(t) " . . e q )(. " . . tFq e 1/?V f RIFT ? Beech ? Low preheat - test 1 Ignition Time 55 m/s kW/m 2 (m/s) -0.5 10 55 0 33.88 20 56 1 169 60 9525 28561 3410 0.0064 0.397 32.87 13.05 12.472 30 58 3 174 90 10100 30276 5260 0.0050 0.404 32.16 13.00 14.142 40 60 5 182 120 11060 33124 7340 0.0032 0.411 29.67 12.19 17.638 50 64 9 193 150 12457 37249 9740 0.0022 0.424 27.17 11.54 21.257 60 69 14 209 180 14633 43681 12660 0.0017 0.441 24.95 11.00 24.608 70 76 21 230 210 17762 52900 16260 0.0012 0.463 22.99 10.64 28.358 80 85 30 255 240 21837 65025 20580 0.0011 0.489 21.03 10.29 30.000 90 94 39 283 270 26877 80089 25660 0.0011 0.514 18.88 9.71 30.836 100 104 49 311 300 32421 96721 31290 0.0011 0.541 16.73 9.05 30.836 110 113 58 344 330 39714 118336 38070 0.0009 0.564 15.06 8.50 34.178 348 120 127 72 387 360 50507 149769 46780 0.0006 0.598 13.89 8.30 41.445 130 147 92 445 390 66979 198025 58290 0.0005 0.643 12.71 8.18 46.969 140 171 116 508 420 86950 258064 71550 0.0005 0.694 11.59 8.04 46.472 150 190 135 10.48 Flame Front Position x Time to position x t (s) time from ignition ?t ?x ?(t 2 ) ?(t) 2 ?(t.x) V f F(t) " . . e q )(. " . . tFq e 1/?V f RIFT ? Beech ? Low preheat - test 2 Ignition Time 55 m/s kW/m 2 (m/s) -0.5 10 55 0 0.394 33.88 20 57 2 172 60 9874 29584 3490 0.0039 0.401 32.87 13.17 15.916 30 60 5 180 90 10818 32400 5460 0.0033 0.411 32.16 13.22 17.321 40 63 8 192 120 12330 36864 7770 0.0021 0.421 29.67 12.49 21.602 50 69 14 211 150 14971 44521 10710 0.0012 0.441 27.17 11.98 28.577 60 79 24 238 180 19102 56644 14490 0.0010 0.472 24.95 11.77 32.416 70 90 35 274 210 25366 75076 19440 0.0008 0.503 22.99 11.57 36.197 80 105 50 315 240 33525 99225 25500 0.0007 0.544 21.03 11.44 38.730 90 120 65 360 270 43650 129600 32700 0.0007 0.581 18.88 10.97 38.730 100 135 80 410 300 56650 168100 41350 0.0006 0.617 16.73 10.31 41.975 110 155 100 468 330 73934 219024 51910 0.0005 0.661 15.06 9.95 46.406 120 178 123 536 360 96918 287296 64800 0.0004 0.708 13.89 9.83 49.004 130 203 148 601 385 121293 361201 77450 0.0004 0.756 12.71 9.61 52.679 135 220 165 0.787 12.71 349 Flame Front Position x Time to position x t (s) time from ignition ?t ?x ?(t 2 ) ?(t) 2 ?(t.x) V f F(t) " . . e q )(. " . . tFq e 1/?V f RIFT ? Beech ? Low preheat - test 3 Ignition Time 60 m/s kW/m 2 (m/s) -0.5 10 60 0 0.411 33.88 20 61 1 184 60 11290 33856 3710 0.0064 0.414 32.87 13.62 12.472 30 63 3 188 90 11786 35344 5670 0.0064 0.421 32.16 13.54 12.472 40 64 4 194 120 12554 37636 7800 0.0046 0.424 29.67 12.59 14.720 50 67 7 202 150 13626 40804 10170 0.0028 0.434 27.17 11.80 18.772 60 71 11 214 180 15306 45796 12930 0.0022 0.447 24.95 11.15 21.257 70 76 16 233 210 18213 54289 16460 0.0013 0.463 22.99 10.64 27.889 80 86 26 257 240 22197 66049 20750 0.0011 0.492 21.03 10.35 30.836 90 95 35 282 270 26622 79524 25530 0.0013 0.517 18.88 9.76 27.568 100 101 41 306 300 31326 93636 30750 0.0013 0.533 16.73 8.92 27.568 110 110 50 334 330 37430 111556 36960 0.0009 0.557 15.06 8.38 33.348 120 123 63 378 360 48254 142884 45710 0.0006 0.588 13.89 8.17 42.292 130 145 85 426 390 61118 181476 55730 0.0006 0.639 12.71 8.12 42.292 140 158 98 480 420 77318 230400 67520 0.0006 0.667 11.59 7.73 40.234 150 177 117 529 450 93929 279841 79710 0.0006 0.706 10.48 7.40 42.448 160 194 134 588 480 116054 345744 94480 0.0005 0.739 9.39 6.94 44.889 170 217 157 642 510 138086 412164 109510 0.0005 0.782 8.33 6.51 43.434 180 231 171 0.806 7.27 Flame Front Position x Time to position x t (s) time from ignition ?t ?x ?(t 2 ) ?(t) 2 ?(t.x) V f F(t) " . . e q )(. " . . tFq e 1/?V f RIFT ? Beech ? Low preheat - test 4 Ignition Time 60 m/s kW/m 2 (m/s) -0.5 10 60 0 0.411 33.88 350 20 61 1 183 60 11165 33489 3680 0.0100 0.414 32.87 13.62 10.000 30 62 2 188 90 11790 35344 5680 0.0046 0.418 32.16 13.44 14.720 40 65 5 194 120 12558 37636 7810 0.0039 0.428 29.67 12.69 15.916 50 67 7 205 150 14043 42025 10330 0.0023 0.434 27.17 11.80 20.817 60 73 13 218 180 15902 47524 13190 0.0018 0.453 24.95 11.31 23.484 70 78 18 237 210 18809 56169 16720 0.0015 0.469 22.99 10.77 25.720 80 86 26 258 240 22316 66564 20800 0.0013 0.492 21.03 10.35 28.284 90 94 34 283 270 26841 80089 25640 0.0012 0.514 18.88 9.71 29.172 100 103 43 312 300 32670 97344 31410 0.0009 0.539 16.73 9.01 32.514 110 115 55 349 330 40995 121801 38670 0.0007 0.569 15.06 8.57 37.544 120 131 71 389 360 50835 151321 46960 0.0007 0.607 13.89 8.43 37.544 130 143 83 431 390 62259 185761 56290 0.0008 0.635 12.71 8.06 36.091 140 157 97 476 420 76074 226576 66970 0.0006 0.665 11.59 7.71 40.775 150 176 116 541 450 98889 292681 81660 0.0004 0.704 10.48 7.37 51.041 160 208 148 611 480 125769 373321 98270 0.0004 0.765 9.39 7.19 51.041 170 227 167 697 503 163437 485809 117196 0.0002 0.799 8.33 6.66 67.198 173 262 202 0.859 8.33 Flame Front Position x Time to position x t (s) time from ignition ?t ?x ?(t 2 ) ?(t) 2 ?(t.x) V f F(t) " . . e q )(. " . . tFq e 1/?V f RIFT ? Beech ? Low preheat - test 5 Ignition Time 60s m/s kW/m 2 (m/s) -0.5 10 60 0 0.411 33.88 20 61 1 183 60 11165 33489 3680 0.0100 0.414 32.87 13.62 10.000 30 62 2 186 90 11534 34596 5600 0.0100 0.418 32.16 13.44 10.000 40 63 3 191 120 12169 36481 7680 0.0046 0.421 29.67 12.49 14.720 50 66 6 200 150 13366 40000 10080 0.0024 0.431 27.17 11.71 20.207 60 71 11 210 180 14726 44100 12670 0.0027 0.447 24.95 11.15 19.272 70 73 13 222 210 16454 49284 15610 0.0027 0.453 22.99 10.42 19.272 351 80 78 18 237 240 18809 56169 19090 0.0015 0.469 21.03 9.86 25.720 90 86 26 261 270 22889 68121 23680 0.0010 0.492 18.88 9.29 30.950 100 97 37 296 300 29574 87616 29870 0.0007 0.523 16.73 8.74 36.952 110 113 53 334 330 37554 111556 37010 0.0007 0.564 15.06 8.50 36.952 120 124 64 380 360 48594 144400 45900 0.0007 0.591 13.89 8.20 39.186 130 143 83 422 390 59850 178084 55170 0.0006 0.635 12.71 8.06 39.703 140 155 95 467 420 73035 218089 65640 0.0008 0.661 11.59 7.66 36.091 150 169 109 510 450 87182 260100 76810 0.0006 0.690 10.48 7.23 39.431 160 186 126 560 480 105182 313600 89960 0.0006 0.724 9.39 6.79 42.448 170 205 145 0.760 8.33 Flame Front Position x Time to position x t (s) time from ignition ?t ?x ?(t 2 ) ?(t) 2 ?(t.x) V f F(t) " . . e q )(. " . . tFq e 1/?V f RIFT ? Beech ? Low preheat - test 6 Ignition Time 60s m/s kW/m 2 (m/s) -0.5 10 60 0 0.411 33.88 20 61 1 183 60 11165 33489 3680 0.0100 0.414 32.87 13.62 10.000 30 62 2 188 90 11790 35344 5680 0.0046 0.418 32.16 13.44 14.720 40 65 5 195 120 12693 38025 7860 0.0033 0.428 29.67 12.69 17.321 50 68 8 204 150 13890 41616 10260 0.0033 0.438 27.17 11.89 17.321 60 71 11 215 180 15441 46225 12980 0.0024 0.447 24.95 11.15 20.207 70 76 16 229 210 17541 52441 16140 0.0018 0.463 22.99 10.64 23.484 80 82 22 246 240 20244 60516 19800 0.0017 0.480 21.03 10.11 24.495 90 88 28 265 270 23493 70225 23980 0.0015 0.498 18.88 9.40 25.520 100 95 35 286 300 27378 81796 28750 0.0013 0.517 16.73 8.65 27.406 110 103 43 312 330 32630 97344 34510 0.0010 0.539 15.06 8.11 30.950 120 114 54 338 360 38246 114244 40740 0.0011 0.567 13.89 7.87 30.246 130 121 61 379 390 48373 143641 49570 0.0006 0.584 12.71 7.42 40.524 140 144 84 432 420 63266 186624 60940 0.0004 0.637 11.59 7.38 47.958 150 167 107 0.686 10.48 352 Flame Front Position x Time to position x t (s) time from ignition ?t ?x ?(t 2 ) ?(t) 2 ?(t.x) V f F(t) " . . e q )(. " . . tFq e 1/?V f RIFT ? Beech ? full preheat - test 1 Ignition Time 841 m/s kW/m 2 (m/s) -0.5 10 33.88 20 32.87 0.00 30 32.16 0.00 40 29.67 0.00 50 27.17 0.00 60 24.95 0.00 70 22.99 0.00 80 842 0 0.999 21.03 21.01 90 847 5 2542 270 2153982 6461764 228890 0.0018 1.000 18.88 18.88 23.484 100 853 11 2556 300 2177754 6533136 255690 0.0021 1.000 16.73 16.73 21.602 110 856 14 2569 330 2199945 6599761 282660 0.0028 1.000 15.06 15.06 18.772 120 860 18 2579 360 2217105 6651241 309550 0.0028 1.000 13.89 13.89 18.772 130 863 21 2588 390 2232594 6697744 336490 0.0039 1.000 12.71 12.71 15.916 140 865 23 2596 420 2246418 6739216 363490 0.0039 1.000 11.59 11.59 15.916 150 868 26 2605 450 2262033 6786025 390820 0.0028 1.000 10.48 10.48 18.772 160 872 30 2617 480 2282937 6848689 418810 0.0022 1.000 9.39 9.39 21.257 170 877 35 2632 510 2309202 6927424 447550 0.0018 1.000 8.33 8.33 23.484 180 883 41 2662 540 2362422 7086244 479410 0.0007 1.000 7.27 7.27 36.914 190 902 60 2706 570 2441534 7322436 514520 0.0005 1.000 6.55 6.55 43.589 200 921 79 2759 600 2537941 7612081 552140 0.0006 1.000 5.83 5.83 41.326 210 936 94 2809 630 2630641 7890481 590200 0.0006 1.000 5.17 5.17 39.377 220 952 110 2859 660 2725241 8173881 629330 0.0006 1.000 4.58 4.58 41.884 230 971 129 2906 690 2815434 8444836 668690 0.0006 1.000 3.99 3.99 39.703 240 983 141 2954 720 2909130 8726116 709250 0.0007 1.000 3.77 3.77 38.267 250 1000 158 1.000 3.54 3.54 353 Flame Front Position x Time to position x t (s) time from ignition ?t ?x ?(t 2 ) ?(t) 2 ?(t.x) V f F(t) " . . e q )(. " . . tFq e 1/?V f RIFT ? Beech ? full preheat - test 2 Ignition Time 841 m/s kW/m 2 (m/s) -0.5 10 20 30 40 50 60 70 80 847 0 1.000 21.03 21.03 90 848 1 2545 270 2159013 6477025 229080 0.0064 1.000 18.88 18.88 12.472 100 850 3 2550 300 2167508 6502500 255040 0.0050 1.000 16.73 16.73 14.142 110 852 5 2555 330 2176013 6528025 281080 0.0064 1.000 15.06 15.06 12.472 120 853 6 2564 360 2191394 6574096 307750 0.0024 1.000 13.89 13.89 20.237 130 859 12 2575 390 2210259 6630625 334850 0.0020 1.000 12.71 12.71 22.509 140 863 16 2589 420 2234339 6702921 362540 0.0025 1.000 11.59 11.59 20.000 150 867 20 2604 450 2260334 6780816 390710 0.0018 1.000 10.48 10.48 23.741 160 874 27 2622 480 2291726 6874884 419660 0.0014 1.000 9.39 9.39 26.458 170 881 34 2647 510 2335701 7006609 450170 0.0011 1.000 8.33 8.33 30.246 180 892 45 2677 540 2389041 7166329 482090 0.0009 1.000 7.27 7.27 33.922 190 904 57 2709 570 2446449 7338681 514920 0.0009 1.000 6.55 6.55 32.514 200 913 66 2741 600 2504561 7513081 548400 0.0010 1.000 5.83 5.83 31.675 210 924 77 2770 630 2557834 7672900 581900 0.0010 1.000 5.17 5.17 31.675 220 933 86 2804 660 2621074 7862416 617110 0.0009 1.000 4.58 4.58 34.178 230 947 100 2847 690 2702387 8105409 655150 0.0006 1.000 3.99 3.99 41.445 240 967 120 2895 720 2794259 8381025 695140 0.0006 1.000 3.77 3.77 41.445 250 981 134 2944 750 2889466 8667136 736290 0.0007 1.000 3.54 3.54 38.086 260 996 149 2992 780 2984602 8952064 778260 0.0006 1.000 3.29 3.29 41.326 270 1015 168 3045 810 3091397 9272025 822530 0.0005 1.000 3.29 3.29 43.589 354 280 1034 187 3112 835 3229350 9684544 866525 0.0003 1.000 2.89 2.89 57.647 285 1063 216 1.000 2.89 2.89 Flame Front Position x Time to position x t (s) time from ignition ?t ?x ?(t 2 ) ?(t) 2 ?(t.x) V f F(t) " . . e q )(. " . . tFq e 1/?V f RIFT ? Beech ? full preheat - test 3 Ignition Time 848 m/s kW/m 2 (m/s) -0.5 10 0 0 20 0 30 0 0 40 0 50 0 0 60 0 70 0 0 80 0 90 0 0 100 0 0 110 848 0 1704 330 1450984 2901912 195940 0.0000 1.000 15.06 15.06 237.777 120 855.5 7.5 2567 360 2195753 6586922 308130 0.0013 1.000 13.89 13.89 27.386 130 863 15 2584 390 2224874 6674472 335950 0.0019 1.000 12.71 12.71 22.980 140 865 17 2606 420 2263878 6791236 364990 0.0011 1.000 11.59 11.59 29.740 150 878 30 2631 450 2307653 6922161 394880 0.0009 1.000 10.48 10.48 34.008 160 888 40 2666 480 2369428 7107556 426780 0.0009 1.000 9.39 9.39 33.212 170 900 52 2704 510 2437600 7311616 459960 0.0007 1.000 8.33 8.33 37.544 180 916 68 2754 540 2528900 7584516 496100 0.0005 1.000 7.27 7.27 43.770 190 938 90 2808 570 2629016 7884864 533900 0.0005 1.000 6.55 6.55 43.770 200 954 106 2872 600 2750360 8248384 574820 0.0005 1.000 5.83 5.83 46.257 210 980 132 2943 630 2888597 8661249 618580 0.0004 1.000 5.17 5.17 52.466 220 1009 161 3021 660 3043505 9126441 665140 0.0004 1.000 4.58 4.58 51.103 355 230 1032 184 3083 690 3168869 9504889 709420 0.0006 1.000 3.99 3.99 41.658 240 1042 194 3137 720 3280757 9840769 753190 0.0006 1.000 3.77 3.77 40.188 250 1063 215 3196 750 3406014 10214416 799490 0.0004 1.000 3.54 3.54 49.666 260 1091 243 3288 775 3606206 10810944 849920 0.0002 1.000 3.29 3.29 70.137 265 1134 286 1.000 3.29 3.29 Flame Front Position x Time to position x t (s) time from ignition ?t ?x ?(t 2 ) ?(t) 2 ?(t.x) V f F(t) " . . e q )(. " . . tFq e 1/?V f RIFT ? Beech ? full preheat - test 4 Ignition Time 841 m/s kW/m 2 (m/s) -0.5 10 33.88 20 30 40 50 60 70 80 90 100 847 0 1696 300 1438210 2876416 178090 0.0000 1.000 16.73 16.73 237.628 110 849 2 2547 330 2162411 6487209 280210 0.0050 1.000 15.06 15.06 14.142 120 851 4 2553 360 2172611 6517809 306400 0.0050 1.000 13.89 13.89 14.142 130 853 6 2565 390 2193131 6579225 333550 0.0018 1.000 12.71 12.71 23.664 140 861 14 2587 420 2231059 6692569 362380 0.0010 1.000 11.59 11.59 31.833 150 873 26 2615 450 2279611 6838225 392450 0.0010 1.000 10.48 10.48 31.833 160 881 34 2649 480 2339315 7017201 424060 0.0009 1.000 9.39 9.39 33.575 170 895 48 2685 510 2403467 7209225 456730 0.0007 1.000 8.33 8.33 37.417 180 909 62 2731 540 2486635 7458361 491900 0.0006 1.000 7.27 7.27 40.104 190 927 80 2783 570 2582419 7745089 529150 0.0005 1.000 6.55 6.55 43.609 200 947 100 2849 600 2706763 8116801 570280 0.0004 1.000 5.83 5.83 49.216 210 975 128 2918 630 2839450 8514724 613270 0.0004 1.000 5.17 5.17 49.666 356 220 996 149 2997 660 2995317 8982009 659850 0.0004 1.000 4.58 4.58 50.759 230 1026 179 3073 690 3149293 9443329 707340 0.0004 1.000 3.99 3.99 52.513 240 1051 204 3152 720 3312902 9935104 756970 0.0004 1.000 3.77 3.77 49.501 250 1075 228 3225 750 3468027 10400625 806730 0.0004 1.000 3.54 3.54 48.990 260 1099 252 1.000 3.29 3.29 357 Appendix 5 Comparison with published results in the literature Plywood Source Test Thickness (mm) " min, . ig q (kW/m 2 ) " . crit q (kW/m 2 ) b (s -0.5 ) t* (s) T ig (?C) h (kW/m 2 K) kpc ((kW/m 2 k 2 ) 2 Radiata Pine ISO ISO 5657 17 13.75 4.6 0.048 432 352 0.041 0.93 Radiata Pine LIFT LIFT 17 16.25 7.2 0.058 293 391 0.044 0.71 Radiata Pine RIFT RIFT 17 16.25 9.9 0.061 273 391 0.043 0.66 Radiata Pine Ngu, CK. 2002 ISO 5657 20 12.00 7.2 0.034 870 321 0.040 1.76 Birch 9-ply Azhakesan et al 1998 Cone 10.7 0.0387 655 386 0.526 Birch faced Jianmin, 1990 Cone 9 14 0.051 383 389 0.8106 "Ordinary" plywood Nisted, 1991 LIFT 12 16.3 0.0369 733 392 1.7871 Japanese plywood - 8.1% MC Fangrat et al, 1996 Cone 18 11.5 328 0.038 0.762 Janssens correlation Japanese plywood -7.7%MC Fangrat et al, 1996 Cone 12 7.8 264 0.033 1.157 Janssens correlation Japanese plywood - 8.2%MC Fangrat et al, 1996 Cone 9 8.5 277 0.039 1.057 Janssens correlation Japanese plywood - 8.2%MC Fangrat et al, 1996 Cone 5.5 9.6 298 0.036 0.877 Janssens correlation US plywood Fangrat et al, 1996 Cone 11.2 14.5 372 0.028 0.334 Janssens correlation Douglas fir (NIST) Fowell 1993 LIFT unspecified 14.7 0.0343 851 372 1.89 Douglas fir (UL) Fowell 1993 LIFT unspecified 17.5 0.0383 681 408 1.77 Douglas fir (Safety eng lab) Fowell 1993 LIFT unspecified 17.0 0.032 974 402 2.51 Round robin test ? all Fowell?s results used the same material but tested at different labs. 358 Douglas fir (NRCC) Fowell 1993 LIFT unspecified 14.0 0.0337 880 363 1.86 LIFT protocol used for results. Douglas fir Dietenberger, 1995 ( c) LIFT 6.55 17 12.1 0.095 335 0.048 Modified protocol & Janssens correlation Douglas fir Dietenberger, 1995 (a) Unspecified unspecified 336 0.254 Douglas fir -5 layers Grexa, White and Dietenberger, 1996 Cone 11.5 13.6 355 0.036 0.236 Janssens correlation Douglas fir -3 layers Grexa, White and Dietenberger, 1996 Cone 12 15.2 369 0.037 0.194 Janssens correlation Oak veneer ply - 5 layers Grexa, White and Dietenberger, 1996 Cone 13 8.9 279 0.030 0.465 Janssens correlation unspecified Quintiere & Harkleroad, 1983 LIFT 6.3-12.7 16 390 0.46-0.54 MDF Source Test Thickness (mm) " min, . ig q (kW/m 2 ) " . crit q (kW/m 2 ) b (s -0.5 ) t* (s) T ig (?C) h (kW/m 2 K) kpc ((kW/m 2 k 2 ) 2 "Customwood" MDF ISO ISO 5657 18 16.25 15.2 0.053 432 352 0.044 0.86 "Customwood" MDF LIFT LIFT 18 16.25 14.7 0.047 293 391 0.044 1.10 "Customwood" MDF RIFT RIFT 18 16.25 15.6 0.053 273 391 0.043 0.86 Ngu, CK. 2002 ISO 5657 20 13.50 12.9 0.043 870 348 0.039 1.04 Asakasan etal 1998 Cone unspecified 7.7 0.046 486 334 0.238 Cleary & Quintiere, 1991 Unspecified unspecified 361 0.732 Henderson, A. 1998 Cone unspecified 1.6 330 1.377 359 Melteca faced MDF Source Test Thickness (mm) " min, . ig q (kW/m 2 ) " . crit q (kW/m 2 ) b (s -0.5 ) t* (s) T ig (?C) h (kW/m 2 K) kpc ((kW/m 2 k 2 ) 2 ISO ISO 5657 18 18.75 2.58 0.043 542 425 0.046 1.47 LIFT LIFT 18 20.00 6.44 0.042 560 440 0.045 1.47 RIFT RIFT 18 21.25 - 1.06 0.051 382 454 0.047 1.07 Melteca faced particle board Source Test Thickness (mm) " min, . ig q (kW/m 2 ) " . crit q (kW/m 2 )) b (s -0.5 ) t* (s) T ig (?C) h (kW/m 2 K) kpc ((kW/m 2 k 2 ) 2 "Regal" brand shelving ISO ISO 5657 18 18.75 2.00 0.044 512 425 0.044 1.27 "Regal" brand shelving RIFT RIFT 18 23.75 10.1 0.057 308 478 0.050 0.97 Asakasan etal 1998 Cone unspecified 16.6 0.0402 576 463 0.575 Cleary & Quintiere, 1991 LIFT unspecified 483 0.804 Hardboard Source Test Thickness (mm) " min, . ig q (kW/m 2 ) " . crit q (kW/m 2 ) b (s -0.5 ) t* (s) T ig (?C) h (kW/m 2 K) kpc ((kW/m 2 k 2 ) 2 ISO ISO 5657 5 11.25 0.59 0.033 944 307 0.039 1.83 LIFT LIFT 5 17.50 2.17 0.052 375 409 0.043 0.88 RIFT RIFT 11.25 2.57 0.03 977 307 0.037 1.67 Azhakesan et al 1998 Cone unspecified 11.1 0.0384 665 392 0.629 Dietenberger, 1995 (b) Cone 6 16.5 393 0.0389 0.277 Dietenberger 1995 (a Cone unspecified 281 0.763 Quintiere & Harkleroad, 1984 LIFT 6.35 10 298 1.87 360 Quintiere & Harkleroad, 1984 LIFT 3.2 14 365 0.88 NZ Beech Source Test Thickness (mm) " min, . ig q (kW/m 2 ) " . crit q (kW/m 2 ) b (s -0.5 ) t* (s) T ig (?C) h (kW/m 2 K) kpc ((kW/m 2 k 2 ) 2 ISO ISO 5657 16 18.00 11.83 0.053 355 415 0.043 0.85 LIFT LIFT 16 18.75 9.24 0.050 395 425 0.044 0.98 RIFT RIFT 16 18.75 12.77 0.034 844 425 0.046 2.30 Ngu, CK. 2002 ISO 5657 20 12.00 11.14 0.025 1618 321.33 0.040 3.27 US Beech Grexa, White Dietenberger, 1996 Cone 15 13.7 358 0.0358 0.508 Radiata Pine (Monterey Pine) Source Test Thickness (mm) " min, . ig q (kW/m 2 ) " . crit q (kW/m 2 ) b (s -0.5 ) t* (s) T ig (?C) h (kW/m 2 K) kpc ((kW/m 2 k 2 ) 2 ISO ISO 5657 16 16.25 12.01 0.039 643 391 0.043 1.54 LIFT LIFT 16 18.75 9.71 0.054 341 425 0.046 0.92 RIFT RIFT 16 18.75 10.93 0.043 531 425 0.046 1.45 Ngu, CK. 2002 ISO 5657 20 15.50 7.78 0.047 460 380.17 0.043 1.08 15% moisture content Moghtaderi, B et al. 1997 Cone unspecified 0.269 In Babrauskas 2003 11% moisture content Henderson, A. 1998 Cone unspecified 10.8 340 0.593 361 Rimu Source Test Thickness (mm) " min, . ig q (kW/m 2 ) " . crit q (kW/m 2 ) b (s -0.5 ) t* (s) T ig (?C) h (kW/m 2 K) kpc ((kW/m 2 k 2 ) 2 ISO ISO 5657 16 18.00 12.75 0.050 397 415 0.045 1.03 LIFT LIFT 16 18.50 9.62 0.050 395 422 0.046 1.06 RIFT RIFT 16 21.25 13.88 0.047 462 454 0.049 1.41 Ngu, CK. 2002 ISO 5657 20 13.50 7.52 0.033 893 348 0.041 1.93 10% MC Henderson, A. 1998 Cone 10.4 355 0.548 In Babrauskas 2003 Heart - 10%MC Henderson, A. 1998 Cone 12.6 355 0.548 In Babrauskas 2003 Macrocarpa Source Test Thickness (mm) " min, . ig q (kW/m 2 ) " . crit q (kW/m 2 ) b (s -0.5 ) t* (s) T ig (?C) ) h (kW/m 2 K) kpc ((kW/m 2 k 2 ) 2 ISO ISO 5657 16 18.00 15.16 0.048 436 415 0.045 1.13 LIFT LIFT 16 18.75 14.68 0.059 288 425 0.046 0.78 RIFT RIFT 16 18.75 15.57 0.056 325 425 0.046 0.89 Ngu, CK. 2002 ISO 5657 20 15.50 12.89 0.055 337 380 0.043 0.79 Henderson, A. 1998 Cone 11.1 402 0.376 Particle board Source Test Thickness (mm) " min, . ig q (kW/m 2 ) " . crit q (kW/m 2 ) b (s -0.5 ) t* (s) T ig (?C) h (kW/m 2 K) kpc ((kW/m 2 k 2 ) 2 Pynefloor Rad. Pine flooring PB Pynefloor ISO 5657 ISO 5657 20 13.75 3.5 0.036 755 352 0.041 1.64 Pynefloor Rad. Pine flooring PB Pynefloor LIFT LIFT 20 18.75 1.7 0.051 386 425 0.044 0.96 Pynefloor Rad. Pine flooring PB Pynefloor RIFT RIFT 20 21.25 4.3 0.063 250 454 0.047 0.70 362 Superflake PB Superflake ISO ISO 5657 20 13.75 3.44 0.032 934 352 0.042 2.10 Superflake PB Superflake RIFT RIFT 20 18.75 5.2 0.051 389 425 0.046 1.05 Dietenberger and Grexa, 1999 cone 257 1.09 7% MC - species unspecified Janssens M, 1992 LIFT 13 20 422 0.0431 0.277 In Babrauskas 2003 Douglas fir Quintiere & Harkleroad, 1984 LIFT 12.7 16 382 0.94 Douglas Fir Dillon, S & Hamins, A. 2003 LIFT 12.8 16 In Babrauskas 2003 Babrauskas and Wetterlund (1999), LIFT 19 10 0.0346 836 319 0.0338 1.215 Cleary & Quintiere. 1991 Unspecified Unspecified 405 0.626 Dietenberger, 1995 (a) Unspecified Unspecified 529.7 1.09 "US particle board" 7.1% MC Fangrat et al, 1996 Cone 12.9 3 145 0.02751 1.8