Effects of surface flame spread of plywood lining on enclosure fire in a modified ISO room

http://jfs.sagepub.com/Journal of Fire Sciences

http://jfs.sagepub.com/content/21/1/67The online version of this article can be found at:

DOI: 10.1177/0734904103021001006

2003 21: 67Journal of Fire SciencesJun Zhang, Fengke Yang, T. J. Shields, G. W. H. Silcock and M. A. Azhakesan

Modified ISO RoomEffects of Surface Flame Spread of Plywood Lining on Enclosure Fire in a

Published by:

http://www.sagepublications.com

can be found at:Journal of Fire SciencesAdditional services and information for

http://jfs.sagepub.com/cgi/alertsEmail Alerts:

http://jfs.sagepub.com/subscriptionsSubscriptions:

http://www.sagepub.com/journalsReprints.navReprints:

http://www.sagepub.com/journalsPermissions.navPermissions:

http://jfs.sagepub.com/content/21/1/67.refs.htmlCitations:

What is This?

- Jan 1, 2003Version of Record >>

by guest on February 19, 2013jfs.sagepub.comDownloaded from

http://jfs.sagepub.com/

http://jfs.sagepub.com/content/21/1/67

http://www.sagepublications.com

http://jfs.sagepub.com/cgi/alerts

http://jfs.sagepub.com/subscriptions

http://www.sagepub.com/journalsReprints.nav

http://www.sagepub.com/journalsPermissions.nav

http://jfs.sagepub.com/content/21/1/67.refs.html

http://jfs.sagepub.com/content/21/1/67.full.pdf

http://online.sagepub.com/site/sphelp/vorhelp.xhtml


Effects of Surface FlameSpread of PlywoodLining on Enclosure Firein a Modified ISO RoomJUN ZHANG* AND FENGKE YANG

Qingdao University of Science and Technology53 Zheng Zhou RoadQingdao, 266042, P.R. China

T. J. SHIELDS, G. W. H. SILCOCK AND

M. A. AZHAKESAN

Fire SERT Center, University of Ulster at JordanstownCo Antrim BT37 0QBNorthern Ireland

(Received April 1, 2002)

ABSTRACT: Surface burning areas of plywood lining installed at walls andceiling of a modified ISO room enclosure were estimated using surfacetemperatures measured in the full-scale enclosure fire tests. Based on the surfaceburning areas, the development of surface flame spread at the lining surface hasbeen assessed experimentally. The correlation of the resulting surface flamespread at different burning times was attempted with other importantexperimental data associated with the enclosure fire growth, including heatrelease, room temperatures, heat fluxes imposed on the floor, the interfaceheight of the hot and cold gas layer etc. Results showed that for the plywoodlining used the surface burning area was a critical factor that significantlyinfluenced the enclosure fire growth and could be reasonably well correlated tomany other parameters of the enclosure fire. The study not only investigated theeffect of surface flame spread of the lining on the enclosure fire growth in-depthbut also produced useful experimental data and analyses for the furtherdevelopment of numerical modeling for enclosure fires lined with combustiblematerials.

*Author to whom correspondence should be addressed.

JOURNAL OF FIRE SCIENCES, VOL. 21 – JANUARY 2003 67

0734-9041/03/01 0067–17 $10.00/0 DOI: 10.1177/073490403032836� 2003 Sage Publications

+ [7.4.2003–8:04am] [67–84] [Page No. 67] REVISED PROOFS I:/Sage/Jfs/Jfs21-1/JFS-32836.3d (JFS) Paper: JFS-32836 Keyword


www.sagepublications.com


KEY WORDS: plywood lining, modified ISO room, surface flame spread, heatrelease, room temperature, interface height.

INTRODUCTION

LINING MATERIALS ARE widely used in buildings to cover walls andceilings. Thus, combustible linings may present potentially high firehazards in buildings, which, once ignited, propagate fire and acceleratethe fire growth in enclosures. Over the last two decades, the firebehavior of building linings has been studied extensively [1–10]. Interms of the ISO room full-scale investigation, the EUREFIC (EuropeanReaction to Fire Classification) program [1] is relevant, in which elevendifferent lining materials were studied with respect to reaction to fireassessment of linings and material classification. In another importantstudy based on the ISO room scale [2], thirteen lining materials wereinvestigated with emphasis on surface flame spread, heat releaseand room fire growth. The experimental data obtained from the twoinvestigations were also utilized by others to develop simulationmodeling of enclosure fire with lining materials [8–10]. However, thedevelopment of a lining fire in an enclosure is closely related toenvironmental conditions such as room temperature, heat fluxes andfluid flows, which vary significantly with the development of the liningfires. The complicated interactions between the lining fire and theresulting fire-induced variations of enclosure environmental conditionsare not fully understood. In particular, there is a need for more relevantexperimental data and correlations.This paper presents some full scale experimental results obtained

from plywood used as a lining in a modified ISO room and attemptsto correlate the surface flame spread estimated by surface temperaturesto the variations of room temperature, heat release, heat fluxes andinterface height of upper and lower gas layers. It also investigatesthe influence of the lining on the enclosure fire development.

EXPERIMENTALS

The Modified ISO Room and Lining Installation

A burn room constructed to ISO standard 9705 [11] and modified toinclude double glazed observation panels in the west wall was used inthe experiments and is shown in Figure 1. The enclosure walls and

68 J. ZHANG ET AL.




ceiling were made of concrete 210mm in thickness and insulated with15mm thick ceramic fiber board. In the case of wall only study, thenorth and east walls were lined with 12mm thick plywood while for thewall and ceiling study case, these two walls and the ceiling partiallycovered with the plywood lining. The remainder of the north and eastwalls were covered with 12mm thick gypsum board while the remainderof the ceiling was left the insulation fiber board covered as detailed inFigures 2 and 3 respectively. Both the plywood and gypsum boardswere conditioned at 23� 2�C and relative humidity of 50� 5% for 74 hbefore the testing.

Plasterboard Plasterboard

2400 3600

1200

1200

100100

200 200

100200100 200 N side E side

1

2

6

7

11

12

16

17

21

22

38131823

49141924

510152025

26

27

28

29

30

31

32

33

34

35

36

37

F1

F2

F4

F3 F8

F10

F9F7

F621161161

22171272

23181383

24191494

252015105

26

27

28

29

30

31

32

33

34

35

36

37

38

39

40

41

42

43

44

45

46

corner line

Thermocouples

F11

Fluxmeters

F5

Figure 2. Locations of surface thermocouples and fluxmeters for the wall only casewith dimensions (mm).

550550

400

Glazing side

E side

N side

Doorway s

ide

PLAN VIEW

E

N Fuel tray50mm fromN & E sides

win

dow

doorwayFuel tray

2400

80

2400

2000

3600

2000

2000

80

80

Figure 1. Schematic diagram of the modified ISO room with dimensions (mm).

Effects of Surface Flame Spread of Plywood Lining 69




Temperature Measurements

The surface temperatures of the lining were measured using insulated1.5mm diameter, K type thermocouples, with the tip of each thermo-couple stuck perpendicularly onto the surface of the lining. The locationsof each of the thermocouples are shown in Figures 2 and 3 respectivelyfor the wall only case and the wall and ceiling case. The thermocouplesused to measure the room corner temperatures and the doorwaytemperatures are shown in Figures 4 and 5 respectively, as suggestedby the ISO 9705 standard [11]. Temperature readings were recordedevery 5 s during tests and temperature data are presented as recorded,i.e. without radiation corrections.

Ignition Source

Mineralized methylated spirits consisting of 93% ethanol andapproximately 5% methanol was used as the fire source in the modifiedISO room fire tests. The fuel was contained in a mild steel fuel tray of

Plasterboard Plasterboard

2400 3600

1200

1200

200 200

100 200

N side E side

1

2

6

7

11

12

16

17

21

22

38131823

49141924

510152025

26

27

28

29

30

31

32

33

34

35

36

37

F2

F3

F5

F4 F10

F9

F8

21161161

22171272

23181383

24191494

252015105

26

27

28

29

30

31

32

33

34

35

36

37

38

39

40

41

42

43

44

45

46

F11

F6

1200

1200

1

Thermocouples

Fluxmeters2

3

4

5

6

7

8

9

10 14 17 19 21 23

11 15 18 20 22 24

12

13

16

Insulation fibre board

Ceiling side200

F1 F7

Figure 3. Locations of surface thermocouples and fluxmeters for the wall and ceilingcase with dimensions (mm).

70 J. ZHANG ET AL.




dimensions 550mm� 550mm� 100mm (length�width�height) andplaced at the north-east corner of the room and 400mm above the floorsurface as shown in Figure 1. That a tray containing liquid fuel waschosen instead of the standard gas burner was to simulate a morerealistic time varying heat release rate, fire source and also the heatfeedback effect of the flame to the fuel surface can be assessed by fuelmass loss variations. For each burn, approximately 8 kg of the fuel wasused and the mass loss was measured by a loadcell installed beneath thetray. Figure 6 shows the heat release measured from two runs for theignition source fuel only, indicating a good repeatability as expected.The fuel generated a heat release that varied from approximately120–170 kW from 80 to 600 s during test, which was measured usingoxygen depletion calorimetry, resulting in an average heat release ofapproximately 145 kW. Since the enclosure fire with lining did notlast more than 400 s, the average value has not included the dataafter 600 s.

Heat Flux Measurement at the Floor

A fluxmeter was installed at the center of the floor area to measurethe total heat flux level imposed on the floor. The results at this location

Figure 4. Locations and positions of the thermocouples (cg) at the room corner withdimensions (mm).





can be used to determine the onset of flashover conditions and will bediscussed in some detail later in this paper.

RESULTS AND DISCUSSION

Surface Temperature Distribution and Surface Ignition

The surface temperatures measured for the plywood lining aremanipulated by a contour plotting program to show isothermal profilesat the surface. The contour graphs at certain critical moments such asinitial ignition and flashover are shown in Figures 7 and 8 respectivelyfor the wall only case and the wall and ceiling case. The times of

145

dg1

dg2

dg3

dg4

dg5

dg6

dg7

dg8

dg9

dg10

900 800 700

2400

2000

1855

1665

1475

1285

1095

905

715

525

335

Bidirectional probes p1-p10

Thermocouples dg1-dg10

p2

p4

p7

p9

p1

p3

p5

p6

p8

p10

doorway

Figure 5. Positions of the thermocouples and bi-directional probes at the doorwaywith dimensions (mm).

72 J. ZHANG ET AL.




occurrence of initial surface ignition recorded via observations are thatfor the wall only case, the north wall ignited first at 100 s and the eastwall at 145 s; for the wall and ceiling case, the north and east wallsignited almost simultaneously at 140 s. The differences in the initialignition times between these two cases may be attributed tothe asymmetrical plume profile of the source fuel fire generatedduring the early stage of the test for the wall only case while asymmetrical plume profile was in the wall and ceiling case. The time toflashover calculated from the initial surface ignition based on the20 kWm�2 criterion [12] at floor is 95 s for the wall only case and 65 s forthe wall and ceiling case. From Figures 7 and 8, the initial ignition areaand location can be identified from the isothermal profiles for each caseif the surface ignition temperature is known. In general, it is difficultto precisely define the surface ignition temperature for solid materialsdue to the complex nature of surface ignition for the materials [13,14].However, an apparent ignition temperature may be defined by theexperimental data of surface temperature due to the fact that once asustained surface ignition occurs, it would result in a sharp increasein temperature. This sharp increase is relatively convenient since itcan be easily identified from the time–surface temperaturecurves. For the purpose of estimating the surface burning area, anapparent surface ignition temperature would be useful.

Figure 6. The heat release measured from two runs for the ignition source fuel.





0 60 120 180 240 300 360 420 480 540 600Width (cm)

0

60

120

180

240

He

ight

(cm

)

0 60 120 180 240 300 360 420 480 540 600Width (cm)

0

60

120

180

240

He

ight

(cm

)

0 60 120 180 240 300 360 420 480 540 600Width (cm)

0

60

120

180

240

Hei

ght

(cm

)

350.0

650

.0

650.0

at 150 sec

0 60 120 180 240 300 360 420 480 540 600Width (cm)

0

60

120

180

240

Hei

ght

(cm

)

350.0

650.0

at 160 sec

0 60 120 180 240 300 360 420 480 540 600Width (cm)

0

60

120

180

240

Hei

ght

(cm

)

0 60 120 180 240 300 360 420 480 540 600Width (cm)

0

60

120

180

240

Hei

ght

(cm

)

350.0

650.0

650.0

at 190 sec

at 100 sec

at 130 sec

at 180 sec

650.0

650.

0

650.0

650.0

650.

065

0.0

650.

035

0.0

350.

0

350.0

Figure 7. Isothermal profiles at the surface of plywood lining for the wall only case atdifferent times during the enclosure fire.

Width ( cm)

0

60

120

180

240

300

360

420

480

Wal

l he

ight

and

cei

ling

dim

ensi

on (

ca)

0 60 120 180 240 300 360 420 480 540 6000 60 120 180 240

5500350.0

350.0

350.0

650.0

650.0 650.0

650.

0

650.0

650.0350.0

350.0

650.

0

350.

0

300 360 420 480 540 600

0 60 120 180 240 300 360 420 480 540 600

Width ( cm)

0

60

120

180

240

300

360

420

480

Wal

l he

ight

and

cei

ling

dim

ensi

on (

cm)

0 60 120 180 240 300 360 420 480 540 600

Wi dth ( cm )

0 60 120 180 240 300 360 420 480 540 600

Wi dth ( cm )

0

60

120

180

240

300

360

420

480

Wal

l hei

ght

and

ceili

ng d

imen

sion

(cm

)

0

60

120

180

240

300

360

420

480

Wal

l hei

ght

and

ceili

ng d

imen

sion

(cm

)

At 140 sec

At 160 sec

At 180 sec

At 200 sec

Figure 8. Isothermal profiles at the surface of plywood lining for the wall and ceilingcase at different times during the enclosure fire.

74 J. ZHANG ET AL.




A typical example is shown in Figure 9 that illustrates the evaluationof the apparent ignition temperature from the surface temperaturesmeasured for the plywood lining. It should be noted that the surfaceignition temperatures obtained are not same and vary at differentlocations. Generally, the highest apparent surface ignition temperatureswere at the initial ignition area where higher energy input may needto cause the initial flaming combustion. A general trend for the ignitiontemperature is found from Figure 10 where it is higher in the regionwhere upward flame spread occurs, and lower in the region wheredownward flame spread takes place. However, most of the ignitiontemperature values fall in the range from 300 to 400�C as shown inFigure 10. With reference to literature [13,15], an average of 350�Cwas used for the estimation of the surface burning area based on theisothermal profiles as shown in Figures 7 and 8.From Figures 7 and 8, it can be seen that the surface flame spread

during the initial stages was relatively slow, corresponding to the regionwhere upward flame spread occurred. After this stage, the surface flamepropagated rapidly to the region where a narrow area between the walltop edge and the ceiling edge, and developed a lateral surface flamespread pattern. This stage was relatively short and therefore lesssignificant in terms of burning area growth. The rapid increase inburning area was at the downward flame spread stage. However, in the

Figure 9. Estimation of apparent surface ignition temperatures from certain typicalsurface temperature profiles.





modified ISO room test with plywood lining installed, the flashovertook place before or around the occurrence of downward flame spread.The variations of surface burning area with time are shown in

Figure 11 where the starting time for both cases are generalized to thetime when the initial sustained surface ignition occurs. It can be seenthat the burning area increases rapidly after a relatively slow stage.Meanwhile, the ceiling installed with plywood accelerates the surfaceflame propagation rapidly and thereby has a greater influence on firegrowth compared to the wall only case.

Effect of Surface Flame Spread on Heat Release

Using the surface burning areas estimated by isothermal profilesat different times, the areas of surface flame spread of the plywoodcan be correlated to the heat release measured by the room calorimeteras shown in Figure 12. Thus, heat release rates from lining materialsbased on full-scale tests can be estimated and used as input datafor modeling development. It should be noted that heat release inFigure 12 includes the contribution of the ignition fuel which willgenerate a higher heat energy after lining surface starts burning due tothe heat feedback to the ignition fuel. However, it was found thatthe effect of feedback was not significant until flashover occurred asshown in Figure 13 in which mass losses of the ignition fuel used for

0

100

200

300

400

500

600

700

800

900

1000

0 50 100 150 200 250 300

Time (sec)

Tem

pera

ture

(o C

)

Figure 10. Estimation of apparent surface ignition temperature from temperatureprofiles.

76 J. ZHANG ET AL.




the control test and the lined tests are compared. Clearly, the sharpdecrease in the fuel mass retention for the lining tests took place afterflashover stage or nearby, indicating the small influence of feedback onheat release at the early stage.From Figure 12, it can be seen that the heat release increased

relatively slowly against surface burning areas until around 3.5m2 ofburning area, which is approximately corresponding to the total areasof the corner burning area and the interface area between the walland ceiling. Heat release increased rapidly when downward flame

Figure 12. Heat release in the enclosure and heat flux at the floor for the plywoodlining fire.

Figure 11. Variations of surface burning areas with the generalized time.





spread occurred and soon after that flashover was achieved, indicatingthe effect of the geometrical configuration of surface lining fire on theISO room enclosure fire growth. Meanwhile, the lined ceiling showsa greater increase in heat release after the flashover stage comparedwith the wall only case, indicating that combustible ceilings wouldcontribute greatly to enclosure fire growth at a later stage.

Effect of Surface Flame Spread onHeat Flux at the Floor

Heat flux levels at the floor area are often used for assessing flashoverstage in enclosure fires [12]. In this study, the total heat flux at the floorcenter was measured by a fluxmeter which recorded total heat fluxlevels. Figure 12 shows the correlation between the surface burning areaand the heat fluxes at the center of the floor for the wall only case andthe wall and ceiling case. As shown in Figure 12, the heat fluxes increaseslowly at the initial stage for both the cases and when the surfaceburning area extends beyond approximately 3.5m2, the heat fluxincreases rapidly. However, it appears that after the flashover pointi.e., 20 kW m�2, the rapid increase in heat fluxes is deterred comparedwith the trend for heat release increase as shown in the same graph. Theslow down at pace is probably because the heat release was measured

Figure 13. Mass loss comparison of the ignition fuel with and without plywood liningin the enclosure fire tests.

78 J. ZHANG ET AL.




outside of the enclosure while the heat fluxes at the floor were measuredinside the enclosure, i.e. the heat fluxes imposed on the floor did notinclude the contribution of the hot gases outside the enclosure.

Effect of Surface Flame Spread on Enclosure Temperatures

Since room temperature is an important parameter for fire growth inenclosures, it was measured at the room corner between the south walland the west wall following the ISO 9705 standard [11]. The location ofthe thermocouple tree at the corner and the position for each thermo-couple is shown in Figure 4. The variations of room temperatures atdifferent levels of heights with surface burning area of plywood lining areshown in Figures 14 and 15 respectively for the wall only case and thewalland ceiling case. It can be seen that temperatures in the upper layerincrease significantlywith the surface burning area and the temperaturesare almost linearly proportional to the surface area for each case.However, as shown in Figures 14 and 15, the wall only case caused ahigher room temperature in terms of surface burning area compared tothe wall and ceiling case. This occurred because in achieving the sameburning area the wall only case took much longer time than the wall andceiling case did. From the Figures 14 and 15, it can be seen that the roomtemperatures in the lower layer increase slightly, showing that surfaceflame spread of plywood lining has little influence on the lower layertemperatures at least until flashover occurs.

0

100

200

300

400

500

600

0 1 2 3 4 5 6

Roo

m te

mpe

ratu

re (C

)

Surface burning area (m )

2100 mm1720 mm1570 mm1420 mm1270 mm970 mm670 mm

Figure 14. Room temperature distribution at different levels of room height for theplywood lining fire in the wall only case.





Effect of Surface Flame Spread on Interface Height

The interface between the upper hot gas layer and the lower coldgas layer in an enclosure fire determines the fluid flowing in and outand is an important parameter for enclosure fire development, especiallyfor zone model. In this study, the interface heights were estimatedbased on the temperature data at various levels of heights in theenclosure as demonstrated in Figure 16 for the control run. Therelationship between the interface heights and the surface burningareas is shown in Figure 17. As shown in Figure 17, the influence ofthe plywood lining on the interface was not significant until flashoveror nearby. With the ceiling partially lined with plywood, the interfaceheight fell rapidly after the occurrence of flashover.

SUMMARY AND CONCLUSIONS

Surface burning areas of plywood lining have been estimated usingthe surface temperature data obtained at different times in the modifiedISO room full-scale fire experiment. Based on the burning areas, theprogress of surface flame spread was assessed experimentally. The study

0

100

200

300

400

500

600

0 2 4 6 8 10 12 14

Surface burning area (m )

Roo

m te

mpe

ratu

re

(C)

2100 mm

1720 mm

1570 mm

1420 mm

1270 mm

970 mm

670 mm

Figure 15. Room temperature distribution at different levels of room height for theplywood lining fire in the wall and ceiling case.

80 J. ZHANG ET AL.




showed that for the plywood lining the spreading of surface flame wasrelatively slow in the corner region close to the ignition fuel, whereupward flame spread occurred. When surface flame extended beyondthis region, the surface flame spread accelerated and significantlyaffected the enclosure fire growth. It was found that surface flamespread was well correlated with many other enclosure fire parameters,including heat release, room temperatures, heat fluxes at the floor, andinterface heights of hot and cold gas layers, indicating that surface flamespread dominated the fire growth of the modified ISO room enclosurelined with plywood. When both wall and ceiling were lined with plywood,

0 2 4 6 8 10 12 14

Surface burning area (m2)

1000

1100

1200

1300

1400

Inte

rfac

e he

ight

(m

m)

ply-wall-interfaceply-wall-ceiling-interface

Figure 17. Variations of the heights of hot–cold gas layer interface with the surfaceburning area of plywood lining during the enclosure fires.

Tem

pera

ture

(C)

200s

650s

800s

50s

0s10

0s

Hei

ght o

f the

roo

m (

mm

2400

2000

1600

1200 80

0400 0

050

100

150

200

250

300

350

Figure 16. The interface heights of the hot–cold gas layer in the enclosure estimatedby room temperatures for the control test.





the combustible ceiling in general accelerated the fire growth. However,it appears that the significant acceleration took place from post-flashover.

ACKNOWLEDGMENT

The authors would like to thank the Fire SERT Center technical stafffor their assistance in the investigation.

REFERENCES

1. Proceedings of EUREFIC Seminar (1991). Interscience CommunicationsLimited, London.

2. Karlsson, B. (1992). Modeling Fire Growth on Combustible Lining Materialsin Enclosures, Report TVBB 1009, Lund University, Sweden.

3. Wickstrom, U. and Goransson, U. (1992). Full-Scale/Bench-Scale Correla-tions of Wall and Ceiling Linings. Fire and Materials, 16: 15–22.

4. Dembsey, N.A. and Williamson, R.B. (1993). The Effect of Ignition SourceExposure and Specimen Configuration on the Fire Growth Characteristics of aCombustible Interior Finish Material, Fire Safety Journal, 21: 313–330.

5. Brehob, E.G. and Kulkarni, A.K. (1998). Experimental Measurements ofUpward Flame Spread on a Vertical Wall with External Radiation, Fire SafetyJournal, 31: 181–200.

6. Yoshida, M., Hasemi, Y., Tanaike, Y., Saito, F. and Fangrat, J. (1999).Comparative Studies in the Reduced-scale Model Box and the Room CornerTest, Interflam’ 99, pp. 23–33.

7. Kerrison, L., Galea, E.R., Hoffmann, N. and Patel, M.K. (1994). AComparison of a FLOW3D Based Fire Field Model with ExperimentalRoom Fire Data, Fire Safety Journal, 23: 387–411.

8. Quintiere, J.G. (1993). A Simulation Model for Fire Growth on MaterialsSubject to a Room-Corner Test, Fire Safety Journal, 20: 313–339.

9. Opstad, K. (1995). Modelling of Thermal Flame Spread on Solid Surfaces inLarge-scale Fires, MTF-Report (1995). 114(D), The University ofTrondheim, Norway.

10. Wade, C.A. (1996). A Room Fire Model Incorporating Fire Growth onCombustible Lining Materials, MSC Thesis, Worcester PolytechnicInstitute, USA.

11. ISO 9705 (1993). Fire Tests-Full-Scale Room Test for Surface Products,International Standards Organisation (ISO), Geneva.

12. Waterman, T.E. (1968). Room Flashover – Criteria and Synthesis, FireTechnology, 4: 25–31.

13. Drysdale, D. (1985). An Introduction to Fire Dynamics, pp. 186–225, JohnWiley and Sons, New York.

82 J. ZHANG ET AL.




14. Thomson, H.E. and Drysdale, D.D. (1987). Flammability of Plastics I:Ignition Temperatures, Fire and Materials, 11: 163–172.

15. Zhang, J., Shields, T.J. and Silcock, G.W.H. (1998). Behavior ofPlywood Lining in Full Scale Room Fire Tests, Journal of Applied FireScience, 8: 3–18.





Effects of surface flame spread of plywood lining on enclosure fire in a modified ISO room

Documents