Fire Protection Ability of Wood Products

Fire protection ability of wood products

Birgit Östman, Rolf Hilling, Lars Boström

weden

SP Technical Research Institute of S of Institute Research Technical SP

SP Report 2012:12

Fire protection ability of wood products

Birgit Östman, Rolf Hilling, Lars Boström

SP Technical Research Institute of Sweden

SP Technical Research Institute of Sweden Box 857, 501 15 Borås, Sweden (headquarters)

SP Report 2012:12 ISSN 0284-5172 ISBN 978-91-87017-28-5

Abstract Wood products may protect underlying products from getting heated and ignited due to their low thermal conductivity and slow charring rate. A European system with K classes for the fire protection ability of coverings has been utilised for wood products. The classes are based on full scale furnace testing and the main parameters are the temperature behind the panel after different time intervals and the ability of the covering to protect underlying products. Three levels are defined: 10, 30 and 60 minutes. Wood- based panels and wood panelling and cladding may fulfil the European K classes for fire protection ability. The criteria for wood products are based mainly on panel thickness. A literature survey shows that similar fire protective behaviour of wood coverings has been verified by different methodologies in several countries in and outside Europe. Results from an extensive test program demonstrate that all K classes may be obtained for wood-based panels (particleboard, plywood, solid wood panels, OSB and hardboard) and for solid wood panelling and cladding. The thickness for achieving each class may vary slightly depending on wood product type and on mounting conditions and fixing device. The end-use applications of the wood products with K class are mainly as wall and ceiling coverings and for protection of underlying materials and structures. K classes are required by building regulations in some countries, e g Germany, Denmark and Sweden.

Key words: fire protection, heat insulation, wood products, structural design, temperature rise

Sammanfattning (Swedish summary) Det europeiska systemet med K-klasser definieras i klassifikationsstandarden EN 13501- 2. Klasserna baseras på ugnsbrandprovning och de huvudsakliga parametrarna är temperaturen bakom skivan efter olika tider och skivans förmåga att skydda bakomliggande material. Tre nivåer har definierats: 10, 30 och 60 minuter. Träbaserade skivor, träpanel och beklädnader kan uppfylla de europeiska K-klasserna för brandskyddande förmåga. Kriterierna för K-klassificering av träprodukter baseras huvudsakligen på skivans tjocklek. En litteraturstudie visar att liknande brandskyddande effekt av träbaserade produkter har verifierats med olika metoder i flera länder inom och utom Europa. Resultat från ett omfattande provningsprogram visar att alla K-klasserna kan uppnås för träbaserade skivor (spånskivor, plywood, massivträskivor, OSB och hård board) och för träpanel. Nödvändig produkttjocklek för att uppnå de olika K-klasserna varierar något för olika typer av träprodukter. Montering med spontade skarvar och fästdon har också viss inverkan. Slutanvändningen av träprodukter med K-klass är främst som vägg- och takbeklädnader.

Preface

The test results reported have been obtained during a long time and within several national and European projects. Public financial support from the European WoodWisdom-Net program and Vinnova Sweden’s Innovation Agency and industry financial support from the European initiative Building with wood (BWW) via CEI-Bois, the Swedish Forest Industries Federation and TMF Trä- och Möbelföretagen is kindly acknowledged. The Nordic companies Swedspan, Forestia, Vänerply and UPM have contributed both financially and by supplying some of the tested products. The research work has been carried out at SP Wood Technology and the fire testing at SP Fire Technology (now SP Fire Research). Special thanks to Rolf Hilling for excellent technical assistance and support. The support, interest and involvement from all partners is kindly acknowledged. The report has been finalized when the approval system according to the CWFT, Classification Without Further Testing, procedure within the European Commission has been completed by publication of the K classes for wood products in Official Journal in December 2014. The classification may now be included in the relevant harmonized product standards and used for CE-marking.

Stockholm January 2015 Birgit Östman SP Wood Technology

Content

Fire protection ability 8

Literature data 9 Conclusion on literature data 14

Fire testing and classification 15 Mounting and fixing 15 Furnace control 16 Measurements and observations 16 Fire classification criteria 17 CWFT – Classification without further testing 17

Wood products tested 18 Joints and fixing devices 18

Fire test results 19

Analyses of fire test data 22

K classes for wood products 24

Final conclusions 26

References 27

Fire protection ability A European system with K classes for the fire protection ability of building panels has been introduced and is defined in EN 13501-2. The K classes are based on full scale furnace testing at horizontal orientation according to EN 14135, and the main parameters are the temperature behind the panel after different time intervals (10, 30 and 60 minutes) and the ability of the covering to protect underlying products. No collapse or falling parts is also required. The test principle is illustrated in Figure 1. The aim with the K classes is to provide fire protection of underlying parts of a structure, e.g. the insulation in a wall or floor element. Two types of K classes are defined depending on the substrate behind. Class K1 10 includes substrates with density less than 3 300 kg/m , while classes K2 10 - K2 60 includes all substrates, so in practise K2 classes should be sufficient for wood-based products. The K classes originate from the Nordic countries, where they have been used mainly for gypsum plasterboards, since the Nordic criteria also include reaction-to-fire requirements. However, in the European system, only the fire protection ability criteria prevail, so this is a great opportunity for wood products to demonstrate their fire protection abilities.

Figure 1. Principle for testing fire protection ability according to EN 14135.

Table 1. European requirements for K classes according to EN 13501-2. Class Test conditions Performance criteria Test Substrate behind Temperature No No damage Time method covering rise behind, collapse on min °C substrate

K1 10 Standard chipboard or < 250 x x 10 EN 14135 substrate ≤ 300 kg/m3

K2 10 10

K2 30 Standard chipboard < 250 x x 30 EN 14135

K2 60 60

Status reports with partial results have been presented at conferences e g /Östman 2010/. Partial results are also included in technical guidelines /Östman et al 2010 and 2012/.

Literature data Several studies have been performed that can be related to the K classes. A brief summary is presented here. Available test evidence is summarised in Figures 2-9 and in Table 2. Further information and data are given in the references. Most of the previous work has been focused on developing methods to calculate the fire resistance of timber frame assemblies and to define the contribution from different assembly components such as panel and insulation products. The work has been performed mainly in model scale furnaces of different sizes. The criteria for temperature increase behind the panel have usually been either 140 or 250 K, but this has been neglected in the present analysis, since the temperature increase at furnace testing is quite rapid at these temperatures and results in only about one minute difference. Data from a study focussing on wood-based products /Norén and Östman, 1985/ are presented in Figure 2. It is obvious that panel thickness is an important factor for the contribution to fire resistance. This property is closely linked to the fire protection ability.

Figure 2. Effect of panel thickness on the contribution to fire resistance of different wood- based panels and gypsum boards /Norén and Östman, 1985/.

A similar study for solid wood panelling has been presented /Norén and Östman, 1986/, see Figure 3. Data on the contribution to fire resistance are given both at joints between the panelling and on panelling between joints, where longer times were observed.

20 on panelling 15 at joints min 10

0 Contribution to fire resistance, to fire Contribution resistance, 0 5 10 15 20 25 30 Thickness, mm

Figure 3. Contribution to fire resistance of solid wood panelling and cladding, both at joints and on panelling /Norén and Östman, 1986/.

Most of the previous work has been focused on developing methods to calculate the fire resistance of timber frame assemblies and to define the contribution from different assembly components such as panel and insulation products. An overview of component additive methods for the calculation of the fire resistance of timber frame structures has been presented /König et al/ and data are summarised in Figure 4. For wood panels (=solid wood panelling and cladding in EN terminology), the low values at tongue and groove (t&g) have been adopted, which explains the difference to wood-based panels.

Figure 4. Basic values for the contribution to fire resistance of some materials used as panels /König et al/.

A similar approach is included in the fire part of Eurocode 5 /EN 1995-1-2/, for fire design of timber structures. The Eurocode approach is summarised below and illustrated in Figure 5.

Partilceboard and fibreboard 30 Plywood

20 Basic value, min

Wood panel (t&g) 10

0 0 10 20 30 40 Thickness, mm

Figure 5. Basic values for the contribution to fire resistance of some materials used as panels according to Eurocode 5 /EN 1995-1-2/.

New supporting data for the Eurocode 5 approach have recently been published /Schleifer and Frangi, Schleifer/.

In France, a different approach has been used to verify fire protection ability. Tests have been performed in a façade arrangement /Hognon/ and based on those, classes for 15 and 30 minutes fire protection have been regulated /Arrêté/. The French data are presented in Figure 6. It is interesting to observe how well these data correspond to the data on contribution to fire resistance of similar wood-based panels presented above.

30 minutes

20 15 minutes Minimum thickness, mm thickness, Minimum

0 300 400 500 600 700 Minimum density, kg/m3

Figure 6. French requirements on fire protection of wood-based panels /Hognon 1987/.

Two Japanese studies have been published /Harada et al, Kikuchi et al/, including data on the fire resistance of thick wood-based boards and the influence of board density, see Figure 7.

Figure 7. Japanese data on plywood (JCPW), particleboard (PB) and MDF /Harada et al/.

Some test data on particleboard and plywood /Hertel,2003 a and b/ are available from Germany, see Figure 8.

Figure 8. German data on particleboard and plywood /Hertel 2003a and b/.

Danish tests /Dahlsgaard Jenssen and Drustrup, Danø and Drustrup/ have been performed with the original Nordic methodology /NT Fire 003/ that is close to the European standard EN 14135. It is shown that 10 minutes protection is reached for 15 mm solid wood panelling. An evaluation report /Bluhme and Drustrup/ shows that the type of tongue and groove joints has minor importance. In the US, thermal barriers are defined in a similar way as the European K classes and a NFPA standard is available /NFPA 275/. Research data for wood products are available /White 1982a and 1982b, White 2003/. Slightly shorter protection times were recorded for substrates made of foam plastic insulation. The most recent data are from Austria /Teibinger and Matzinger/ and includes information on particleboard, hardboard and OSB, see Figure 9.

Figure 9. Austrian data on particleboard, hardboard and OSB /Teibinger and Matzinger/.

Table 2. Summary of literature data. Minimum thickness to reach criteria for 10 or 30 minutes protection, mm Wood-based panels Wood-based panels Solid wood Reference Density < 400 kg/m3 Density > 400 kg/m3 panelling Density > 400 kg/m3 10 30 WBP 10 30 WBP incl 10 30 min min incl min min min min PB, PW, HB, Norén and Östman 13 softboard 6-9 22-25 8 MDF 1985 Norén and Östman 11/15 at joint 1986 11 32 PW 9 28 PB, PW, FB 18 König et al at joint 10 32 PW 9 27 PB, PW, FB 20 EN 1995-1-2 at joint 30 general Lignum 30 PB, PW, MDF Harada et al < 13 PB, PW Hertel 2003a < 24 PB, PW Hertel 2003b 15 Danø and Drustrup Dalsgaard-Jensen 14/15 and Drustrup Teibinger and < 10 PB, PW, OSB Matzinger 15 PW, PB, HB 15 White 1982a 11 PW White 1982b PW, OSB, 29 White 2003 LVL Kikuchi and 13 31 Komazawa WBP = wood-based panels; PW = plywood; PB = particleboard; HB = hardboard; FB = fiberboard; MDF = medium density fiberboard; OSB = oriented strand board; LVL = laminated veneer lumber.

Conclusion on literature data The literature data are summarized in Table 2. It is obvious that panel thickness is the most important factor for the contribution to fire resistance for wood-based panels. Density has a minor effect. Similar thickness of the wood products to reach 10 and 30 minutes fire protection has been demonstrated. These data are promising for reaching European K class, but the methodologies used have been different, so further testing according to the European standards is needed to reach the European K classes.

Fire testing and classification The wood products have been tested in a horizontal furnace according to EN 14135 Coverings - Determination of fire protection ability. Different thicknesses of covering, design and geometry for joints and fixing methods of the products are included. At each test, two full size specimen 3,0 x 2,4 m were included, since the furnace size is 5 m x 3 m. A series of pretesting with smaller elements 1,0 x 2,4 m was also performed.

Mounting and fixing All products were mounted on the substrate prescribed in EN 14135, 19 mm particleboard 680 kg/m3, without any air gap or cavity behind. The coverings were fixed to the substrate in the same way as in practise, i.e. with screws for all wood-based panels except hardboards/medium board where brads were used, and with nails for solid wood panelling and cladding. All elements consisted of segments of the tested covering with joints according to EN 14135. The supporting system consisted of a framework of wooden beams with cc 600 mm and the particleboard substrate was mounted on the lower side of the framework. Further details on the mounting and fixing are given in chapter Wood products tested and in the test reports specified in Tables 5 and 6. Typical design of test elements and instrumentation of full size specimen is given in Figure 10 a. Design of the elements for pretesting is given in Figure 10 b.

Figure 10 a. Typical design of segments in a full size tested covering 3,0 x 2,4 m with joints according to EN 14135. Thermocouples behind the tested cladding are included to the left and typical screw spacing to the right (screw size and edge distance not to scale).

Figure 10 b. Pretest element with six different specimen 1,0 x 2,4 m tested at the same time. Joints were included and three thermocouples were used for each specimen.

Furnace control The furnace was controlled in accordance with EN 1363-1. The furnace temperature was measured with 5 plate thermometers and the measuring junctions were positioned approximately 100 mm below the fire exposed surface at the commencement of the test. The average temperature and the temperature of each plate thermometers were recorded and reported in the test reports. The pressure in the furnace in relation to the ambient pressure in the test hall was measured and controlled 100 mm below the fire exposed surface of the test specimen. The furnace was controlled to an overpressure of approximately 20 Pa.

Measurements and observations The temperature rise on the lower side of the particleboard substrate was measured with 8-9 thermocouples for the full size elements and recorded. Photographs were taken before and after the test. After the test, the tested covering was inspected for collapse and the particleboard substrate for damage/charring.

Fire classification criteria There are two types of K classes defined in EN 13501-2 depending on the substrate 3 behind. Class K1 10 includes substrates with density less than 300 kg/m , while classes K2 10 - K2 60 includes all substrates, so in practise K2 classes should be sufficient for wood- based products. The classification criteria for the K classes are defined in EN 13501-2.

A covering designated K2 is considered to give the prescribed protection for materials behind the covering if during a test in accordance with EN 14135 within the classification period (10 min, 30 min or 60 min) there is no collapse of the covering or parts of it and also if the following requirements are fulfilled. For a covering without a cavity or cavities behind it  during the test the mean temperature measured on the lower side of the substrate shall not exceed the initial temperature by more than 250 °C and the maximum temperature measured at any point of this side shall not exceed the initial temperature by more than 270 °C, and  after the test there shall be no damage or charred material at any point of the substrate.

CWFT – Classification without further testing The classification of the fire protection ability of the wood products tested has been performed according to the CWFT - Classification without further testing procedure. This procedure may be used for products which have been proven to be stable in a given European class (on the basis of testing to the appropriate EN test methods within the scope of their variability in manufacturing allowed by the product specification (harmonised standard or ETA European Technical Approval)). CWFT is a list of generic products, not a list of proprietary products. The CWFT procedure is described in a document from DG Enterprise /CONSTRUCT 01/491 rev 3, 2004/. Products claiming CWFT must be clearly above the lower class limits, to provide a safety margin. The European Standing Committee on Construction, SCC, makes the final decision based upon the recommendations from the European Group of Fire experts, EGF. The advice of the EGF will largely determine whether the request is forwarded to the SCC for opinion. Wood products are good examples of products having a stable fire performance. The CWFT approach has earlier been applied for the reaction to fire performance of several wood products. Results have been published e g /Östman and Mikkola 2006/. This is the first CWFT case dealing with fire protection ability performance.

Wood products tested Different types of wood products according to the harmonised standards EN 13986 and EN 14915 have been tested, five types of wood-based panels (plywood, particleboard, oriented strand board (OSB), hardboard/medium board and solid wood panels (SWP)) and solid wood panelling and cladding (SWPC), see Table 3. The relevant product standards referred to are EN 300 Oriented Strand Boards (OSB) EN 312 Particleboards — Specifications EN 622-2 Fibreboards - Specifications - Part 2: Requirements for hardboards EN 622-3 Fibreboards - Specifications - Part 3: Requirements for medium boards EN 636 Plywood — Specifications EN 13353 Solid wood panels (SWP) – Requirements (multi layer products) EN 14519 Solid wood panelling and cladding (SWPC) – softwood machined profiles with tongue and groove EN 14951 Solid hardwood panelling and cladding (SWPC) – Machined profiles elements

Table 3. Wood products tested Wood EN harmonized Thickness Fixing device Joints product product standard mm (nail, screw etc)

Particleboard EN 13986 10, 12, 22, 25 screw square edges, t&g Plywood EN 13986 9, 12, 24 screw square edges, t&g Hardboard EN 13986 9 brad square edges OSB EN 13986 10, 12, 25, 30 screw square edges, t&g SWP EN 13986 13, 26, 52 screw square edges, t&g SWPC EN 14915 15, 27 nail tongue & groove

Joints and fixing devices The joints within a wood element were either with straight square edges or with tongue and groove profiles without gaps. Typical design is given in Figure 11. The thickness at the joints was the same as for the wood product.

Figure 11. Typical design of joints with square edges (to the left) or with tongue and groove (to the right).

The fixing devices were chosen according to industry guidance /Guide on screws and nails/ and producers´ recommendations, see Table 4. The edge distance was 3d (nominal diameter) for wood-based panels and 5d for SWPC.

Table 4. Fixing devices used for wood products Fixing Length/diameter(d), mm Spacing at Used for device edge, mm Screw 25/2,5; 30/3.5; 30/3,9; 30/4,2; 41/4,2; 200 Particleboard, OSB, 50/3,5; 57/4,2; 75/4,1 plywood, SWP Brad 40/1,7 100 Hardboard Nail 50/2,0; 50/3,5; 60/2,3 600 SWPC

Fire test results Typical examples of pictures from the fire testing according to EN 14135 are given in Figure 12.

Figure 12. Example from testing solid wood panelling. Above the exposed side of test sample before and after fire testing. Below to the left: The substrate behind the test sample after the fire testing. Below to the right: Measured temperatures behind the test sample.

The pretest results with smaller specimen 1,0 x 2,4 m are summarized in Table 5 and illustrated Figure 13. They were considered to be promising, so standard testing with full size elements started. The standard test results with full size specimen are summarized in Table 6.

Table 5. Pretesting according to EN 14135 with smaller elements (1,0 x 2,4 m)

Thick Density Joints Fixing device (nail, screw etc), End Max Time to Char Pass SP Test -ness kg/m3 t&g mm tim temp Δ temp at Test nr mm e rise at > 250 K substr Repor Product Lengt Spacing min end min ate t Type Ø h at in time C * edge field Particleboard 10 ca 600 x screw 30 3,9 200 300 10 112 - No Yes -“- 10 ca 600 - -“- 30 3,9 200 300 10 112 - No Yes Particleboard 12 ca 600 x -“- 30 3,9 200 300 10 81 - No Yes 1 A -“- 12 ca 600 - -“- 30 3,9 200 300 10 90 - No Yes OSB 10 ca 600 - -“- 30 3,9 200 300 10 158 - No Yes -“- 12 ca 600 x -“- 30 3,9 200 300 10 132 - No Yes Plywood 12 ca 400 - screw 30 3,9 200 300 10 121 - No Yes -“- 12 ca 400 - -“- 30 3,9 200 300 10 128 - No Yes Plywood 9 ca 400 - -“- 30 3,9 200 300 10 194 - Yes No 2 B Hardboard 9 ca 700 - -“- 30 3,9 200 300 10 125 - No Yes SWPC 15 ca 400 x nail 60 2,3 600 - 10 90 - No Yes -“- 15 ca 400 - -“- 60 2,3 600 - 10 88 - No Yes Particleboard 22 611 - screw 57 4,2 200 300 30 >250 26,5 Yes No 2 x 57 4,2 200 300 Particleboard 592 - -“- 30 >250 22,8 Yes No 12 Plywood 24 653 - -“- 57 4,2 200 300 30 176 - Yes No 3a C 2 x 57 4,2 200 300 Plywood 699 - -“- 30 >250 20,7 Yes No 12 OSB 25 615 - -“- 57 4,2 200 300 30 >250 29,5 Yes No SWP 27 454 x nail 60 2,3 600 - 30 98 - Yes No Plywood 24 487 x screw 41 4,2 200 300 30 172 - Yes No OSB 25 611 x -“- 41 4,2 200 300 30 >250 28 Yes No Particleboard 22 735 x -“- 41 4,2 200 150 30 >250 27 Yes No 2 x 3 b Plywood 509 x -“- 41 4,2 200 150 30 >250 26 Yes No H 12 2 x Particleboard 643 x/- -“- 41 4,2 200 150 30 >250 25 Yes No 12 SWPC 27 434 x nail 60 2,3 600 - 30 119 - No Yes

OSB = Oriented Strand Board SWPC = Solid wood panelling and cladding, planed tongue and groove with equal thickness

*/Hilling and Boström/

Figure 13. Pretesting results for tests without char at substrate.

Table 6. Standard testing (3,0 x 2,4 m elements) according to EN 14135

Thick- Density Joint Fixing device (nail, screw etc), mm End Max Time to Char Pas SP 3 Tes ness kg/m t&g time temp Δ temp at s Test t nr min rise at > 250 K subst repor Product mm Lengt Ø Spacing Type end min t h at in edge time * field C 4b Particleboard 10 629 - screw 30 3,9 200 300 10 148 - Yes No E 4c Particleboard 10 652 x -“- 30 3,9 200 150 10 109 - No Yes G1 4c Particleboard 12 583 - -“- 30 3,9 200 300 10 92 - No Yes G2 5 Hardboard 9 771 - brad 40 1,7 100 200 10 157 - No Yes F1 6 OSB 10 590 - screw 30 4,2 200 300 10 178 - No Yes I1 4a Plywood 12 531 - -“- 30 3,9 200 300 10 120 - No Yes D 4d Plywood 12 477 - 2 screws 30 3,9 200 300 10 129 - No Yes L1 4d SWP 13 466 - -“- 30 3,9 200 300 10 107 - No Yes L2 F2 15 625 - nail 60 2,3 200 - 10 90 - Yes No 5 SWPC rev1 6 SWPC 15 457 x -“- 60 2,3 600 - 10 98 - No Yes I2 2 22 616 - 50 3,5 200 300 30 573 29/ Yes No N1 7 Particleboard screws 10 Particleboard 22 683 - -“- 50 3,5 200 300 30 280 29 Yes No R1 2 25 660 x 50 3,5 200 300 30 114 >30 No Yes T1 12 Particleboard screws 12 Particleboard 2x15 620 x -“- 25 2,5 200 300 30 788 26 Yes No T2 2 25 553 x 41 4,2 200 300 30 208 >30 Yes No K2 9 OSB screws 9a OSB 25 609 x -“- 50 3,5 200 300 30 172 >30 Yes No O2 R2 30 600 - -“- 50 3,5 200 300 30 96 >30 No Yes 10 OSB rev1 8 Plywood 24 467 x -“- 50 3,5 200 300 30 482 >30 Yes No M2 2 2 x 12 467 - 25 2,5 200 300 30 820 19 Yes No M1 8 Plywood screws 9a Plywood 24 467 x -“- 50 3,5 200 300 30 230 >30 No Yes O1 7 SWP 26 447 x -“- 50 3,5 200 300 30 115 >30 No Yes N2 K1 27 454 x nail 60 2,3 600 - 30 114 >30 No Yes 9 SWPC rev1 13 Plywood 2 x 24 465 x -“- 50 3,5 200 300 60 900 48 Yes No S2 14 Plywood 2 x 27 477 x -“- 50 3,5 200 300 60 851 56 Yes No U2 2 P2 52 492 - 75 4,1 200 300 60 102 >60 Yes No 11 SWP screws rev1 at 53 443 -“- 75 4,1 200 300 60 101 >60 Yes No U1 14 SWP unexp 15 SWP 52 443 x -“- 75 4,1 200 300 60 102 >60 Yes No V1 at V2 52 443 -“- 75 4,1 200 300 60 95 >60 No Yes 15 SWP unexp rev1 2 2 x 26 472 - 50 3,5 200 300 60 226 >60 Yes No S1 13 SWP screws P1 2 x 27 454 x nail 50 2,0 600 - 60 96 >60 No Yes 11 SWPC rev1

OSB = Oriented Strand Board SWP = Solid wood panels, multi layer SWPC = Solid wood panelling and cladding, planed tongue and groove with equal thickness */Hilling et al/

Analyses of fire test data

The fire test results have been analysed in terms of panel thickness, type of wood product, joints and char at substrate. The results are illustrated in figures 14-16. Most tested products with no char at substrate also passed the temperature rise criterion on < 250oC at substrate. But some of the products passing the temperature criterion had some char at the substrate behind the covering. This may be caused by heat exposure at joints or fixing devices. However, no difference between straight and t&g joints could be observed, see lower diagram in figure 14. The conclusion is that the combination of requirements on maximum temperature on substrate and lack of charring is useful as test criteria.

No char at substrate

Figure 14. Full scale test results. Above for wood products with no char at substrate behind the covering and below for products with char at substrate.

Protection time, min

Panel thickness, mm

Figure 15. Protection time according to EN 14135 for wood-based panels and solid wood panelling with different thickness.

The protection time for wood-based panels with different thickness is illustrated in Figure 15. The dependence on panel thickness is clear. The real protection time is longer, but not recorded, since the tests had to be terminated at the exact end times according to EN 14135 in order to be able to observe possible charring at the substrate. The pretesting and the full size data have also been compared, see Figure 16. It is obvious that the results are quite similar, especially at 10 minutes fire exposure for which most of the tests were performed.

Figure 16. Comparison of testing wood products in model scale and full scale according to EN 14135. To the left for 10 minutes fire exposure and to the right for 30 minutes.

K classes for wood products K class has been verified for fifteen wood products according to the CWFT procedure. Detailed documentation was supplied to the European Commission services /CWFT case report for K classes/. The documentation was carefully checked in several steps and at different levels and finally approved by SCC, the European Standing Committee on Construction. In all cases the criteria according to EN 13501-2 were fulfilled:  The maximum temperature rise behind the covering was less than 250 oC  There was no char on the substrate behind the covering The safety margin to the class limit is generally very large and in average < 100 oC.

Table 7. Classification of wood products based on test results according to EN 14135 Thick- Density Joint Fixing device (nail, Max Char Class SP ness kg/m3 screw etc) temp at EN Report part no Product mm Min Max rise at substr 13501- * length spacing end ate 2 Type mm mm time C a Plywood 12 531 square edges screws 30 200 120 No K2 10 D a Plywood 12 477 square edges screws 30 200 129 No K2 10 L1 a Hardboard 9 771 square edges brads 40 100 157 No K2 10 F1 tongue & a Particleboard 10 652 groove screws 30 150 109 No K2 10 G1 a Particleboard 12 583 square edges screws 30 200 92 No K2 10 G2 a OSB 10 590 square edges screws 30 200 178 No K2 10 I1 a SWP 13 466 square edges screws 30 200 107 No K2 10 L2 tongue & 60 600 a I2rev1 SWPC 15 457 groove nails 98 No K2 10 tongue & 50 200 O1 Plywood 24 467 groove screws 230 No K2 30 tongue & 50 200 T1 Particleboard 25 660 groove screws 114 No K2 30 t&g at short OSB 30 600 sides only screws 50 200 96 No K2 30 R2 tongue & 50 200 N2rev1 SWP 26 447 groove screws 115 No K2 30 tongue & 60 600 K1rev1 SWPC 27 454 groove nails 114 No K2 30 tongue & 75 200 V2rev1 SWP 52 443 groove screws 95 No K2 60 tongue & 50 600 P1rev1 SWPC 2x27 454 groove nails (in each 96 No K2 60 layer)

a 3 Fulfils also K110 for substrates ≥ 300 kg/m * Parts of SP Test Report P905327 /Hilling et al/

OSB = Oriented Strand Board SWP = Solid wood panels, multi layer SWPC = Solid wood panelling and cladding, planed tongue and groove with equal thickness

Table 8. Table in Commission Decision /Official Journal/ CLASSES OF FIRE PROTECTION ABILITY PERFORMANCE FOR WOOD-BASED PANELS and SOLID WOOD PANELLING AND CLADDING FOR WALL AND CEILING COVERING Product 1) EN product Product detail 2) Minimum Minimum K Class standard mean thickness 3) density (mm) (kg/m3) With and without 4) Hardboard EN 13986 5) 800 9 K 10 tongue and groove 2 With and without 4) OSB EN 13986 6) 600 10 K 10 tongue and groove 2 With tongue and 4) Particleboard EN 13986 7) 600 10 K 10 groove 2 With and without 4) Particleboard EN 13986 6) 600 12 K 10 tongue and groove 2 With and without 4) Plywood EN 13986 6) 450 12 K 10 tongue and groove 2 Solid wood With and without 4) EN 13986 6) 450 12 K 10 panels tongue and groove 2 With tongue and Particleboard EN 13986 8) 600 25 K 30 groove 2 With tongue and OSB EN 13986 8) 600 30 K 30 groove 2 With tongue and Plywood EN 13986 8) 450 24 K 30 groove 2 Solid wood With tongue and EN 13986 8) 450 26 K 30 panels groove 2 Solid wood With tongue and EN 13986 9) 450 52 K 60 panels groove 2 Solid wood With tongue and 4) panelling and EN 14915 10) 450 15 K 10 groove 2 cladding Solid wood With tongue and panelling and EN 14915 10) 450 27 K 30 groove 2 cladding Solid wood With tongue and 12) panelling and EN 14915 11) 450 2 x 27 K 60 groove 2 cladding 1) Mounted directly on any substrate without an air gap 2) Joints with square edges or tongue and groove profile and with the same thickness as the wood product and without gaps 3) Class as set out in Decision 2000/367/EC 3 4) K1 10 for substrates ≥ 300 kg/m 5) Brad length minimum 40 mm and spacing maximum 100 mm 6) Screw length minimum 30 mm and spacing maximum 200 mm 7) Screw length minimum 30 mm and spacing maximum 150 mm 8) Screw length minimum 50 mm and spacing maximum 200 mm 9) Screw length minimum 75 mm and spacing maximum 200 mm 10) Nail length minimum 60 mm and spacing maximum 600 mm 11) Nail length minimum 50 mm (in each layer) and spacing maximum 600 mm 12) The two layers are mounted with the longitudinal direction of the layers perpendicular to each other.

Final conclusions Main product parameters influencing the fire protection ability of wood-based panels and solid wood panelling and cladding are thickness and density. The joints and the fixings to the substrate are also important. Wood-based panels of types plywood, particleboard, OSB and solid wood panels (according to EN 13986) with thickness at least 10 mm and hardboard/medium board 3 with thickness at least 9 mm fulfil classes K1 10 (for substrates ≥ 300 kg/m ) and K2 10. Wood-based panels of types plywood, particleboard, OSB and solid wood panels with thickness at least 24-30 mm fulfil class K2 30. Solid wood panels with thickness at least 52 mm fulfil class K2 60. Solid wood panelling and cladding (according to EN 14915) planed with tongue and groove and with equal thickness at least 15 mm fulfil classes K1 10 (for substrates ≥ 300 3 kg/m ) and K2 10. Solid wood panelling and cladding of least 27 mm fulfil class K2 30 and of at least 2 x 27 mm class K2 60. The classification has been approved by the European CWFT procedure and published in December 2014 /Official Journal /. Wood based products fulfilling the different K classes are summarised in Table 9. These results may be included in the relevant harmonised product standards and used for CE-marking. The end-use applications of the wood products tested are mainly as wall and ceiling coverings.

Table 9. Wood based products fulfilling K classes for fire protection ability. K Product EN Joints Fixing device Min. Min. class standard Type Min Max density thickness length spacing kg/m3 mm mm at edge mm Tounge Particleboard EN 13986 screw 30 150 600 10 & groove -“- EN 13986 - screw 30 200 600 12 Plywood EN 13986 - screw 30 200 450 12 K210 a screw 30 200 OSB EN 13986 - 600 10 Hard board EN 13986 - brad 40 100 800 9 Solid wood panel EN 13986 - screw 30 200 450 13 Solid wood panelling and Tounge EN 14915 nail 60 600 450 15 cladding & groove Particleboard EN 13986 -“- screw 50 200 600 25 Plywood EN 13986 -“- screw 50 200 450 24 OSB EN 13986 -“- screw 50 200 600 30 K230 Solid wood panel EN 13986 -“- screw 50 200 450 26 Solid wood panelling and EN 14915 -“- nail 60 600 450 27 cladding Solid wood panel EN 13986 -“- screw 75 200 450 52 60 K260 Solid wood panelling and EN 14915 -“- nail in each 600 450 2 x 27 cladding layer a 3 Fulfills also K110 for substrates with density ≥ 300 kg/m behind the cladding.

References Arrêté du 6 octobre 2004 portant approbation de dispositions complétant et modifiant le règlement de sécurité contre les risques d’incendie et de panique dans les établissements recevant du public, Journal officiel de la République Française, 29 décembre 2004. Bluhme D and Drustrup A. DBI Evaluation report PH12869 (in Danish), 2007-01-04, Danish Institute of Fire and Security Technology. Classification of products of known and stable performance - Procedural aspects. CONSTRUCT 01/491 rev 3, 2004. CWFT case report for K Classes, CWFT doc 174 rev 4, 2012 and Amendment, 2013. Dahlsgaard-Jenssen T and Drustrup A: DBI Test report PG11651 (in Danish), 2006-07-27, Danish Institute of Fire and Security Technology. Danø E and Drustrup A. DBI Test report PG11418 (in Danish), 2005-01-05, Danish Institute of Fire and Security Technology. EN 1995-1-2:2004. Eurocode 5 – Design of timber structures. Part 1-2: General rules – Structural fire design. EN 13501-2. Fire classification of construction products and building elements - Part 2: Classification using data from fire resistance tests, excluding ventilation services. European Standard. EN 13986. Wood-based panels for use in construction - Characteristics, evaluation of conformity and marking. European Standard. EN 14135. Coverings – Determination of fire protection ability. European Standard. EN 14915 Solid wood panelling and cladding – Characteristics, evaluation of conformity and marking. European Standard. Guide on screws and nails. (Skruv- och spikguide in Swedish), Swedish Forest Industries Federation, 2010. Harada T, Uesugi S and Masuda H. Fire resistance of thick wood-based boards, J Wood Sci, Vol 52, 544-551, 2006. Hertel C. Orientierende Prüfung in Anlehnung an DIN 4102 Teil 2 und DIN EN 1362 Teil 1 von 13 mm dicken Spanplatten und 15 mm dicken Sperrholzplatten… Prüfbericht MFPA Leipzig Nr PB III/B -03-107, 2003a. Hertel C. Orientierende Prüfung in horizontaler Einbaulage in Anlehnung an DIN 4102 Teil 2 und DIN EN 1362 Teil 1 von 22 mm dicken Spanplatten und 24 mm dicken Sperrholzplatte. Prüfbericht MFPA Leipzig Nr PB III/B -03-160, 2003b. Hilling R and Boström L. Development fire test of six coverings. SP Test Reports P905327, parts 01A, 01B, 01C and 01H, 2009-2010. Hilling R et al. Fire test of a covering. SP Test Reports P905327. Totally 30 parts 01D-01V2, 2009-2012. Hognon B. Mesure des nuisances principales dues au feu, Service sécurité au feu, CSTB activités 1986, Extrait du cahier du CSTB 2175-1 Septembre 1987. Kikuchi S and Komazawa K. Thermal Insulation and Fire Integrity Property of Various Thickness Wood Panel. J Hokkaido For Prod Res Inst, Vol 15, No 1, 2001.

König J, Oksanen T and Towler K. A review of component additive method used for the determination of fire resistance of separating light timber frame structures, CIB W18 Paper 33-16- 3, 2000. Lignum-Dokumentation Brandschutz 4.1 Bauteile in Holz - Decken, Wände und Bekleidungen mit Feuerwiederstand. SIA ETH, Empa and Lignum, 2007. NFPA 275 Standard Method of Fire Tests for the Evaluation of Thermal Bariers Used over Foam Plastic Insulation, 2009. Norén J and Östman B. Contribution to fire resistance from building panels. Proc First International Symposium Fire Safety Science 325-335, Washington, 1985. Norén J and Östman B. Contribution to fire resistance from wood boarding (in Swedish), TräteknikCentrum Rapport 8609054, 1986. NT Fire 003. Nordtest method 003, Coverings: Fire protection ability, 1985. Official Journal. COMMISSION DELEGATED REGULATION (EU) No 1291/2014 of 16 July 2014 on the conditions for classification, without testing, of wood-based panels under EN 13986 and solid wood panelling and cladding under EN 14915 with regard to their fire protection ability, when used for wall and ceiling covering. Official Journal of the European Union. L 349/25. 5.12.2014. Schleifer V and Frangi A. Separating function of light timber frame assemblies, World Conference on Timber Engineering, WCTE, 2006. Schleifer V. Zum Verhalten von Raumabschlissenden Mehrschichtigen Holzbauteilen in Brandfall. Doktorsabhandlung, ETH Zürich, 2009. Teibinger M and Matzinger I. Basic Principles of Classifying the Fire Resistance of Timber Constructions. HFA Report 416, Holzforschung Austria, 2008. White R H. Wood-based Paneling Thermal Barriers. Forest Products Laboratory, Research Paper FPL 408, 1982a. White R H. Effect of calcium silicate substrate on thermal barrier fire testing. Forest Products J Vol 32, 7, 1982b. White R H. Fire Resistance of Engineered Rim Board Products. Forest Products Laboratory, Research Paper FPL-RP-610, 2003. Östman B and Mikkola E. European classes for the reaction to fire performance of wood products. Holz als Roh- und Werkstoff 64:327-337, 2006. Östman B. Fire protection ability of wood products – A status report. Interflam 2010: Proceedings of the twelfth international conference, 71-81, London, UK, Interscience Communications, 2010. Östman et al. Fire safety in timber building – Technical guideline for Europe. SP Report 2010:19, 2010. Östman et al. Brandsäkra trähus 3 – Nordisk-baltisk kunskapsöversikt och vägledning (in Swedish). SP Rapport 2012:18, 2012.

SP Technical Research Institute of Sweden Our work is concentrated on innovation and the development of value-adding technology. Using Sweden's most extensive and advanced resources for technical evaluation, measurement technology, research and development, we make an important contribution to the competitiveness and sustainable development of industry. Research is carried out in close conjunction with universities and institutes of technology, to the benefit of a customer base of about 10000 organisations, ranging from start-up companies developing new technologies or new ideas to international groups.

SP Technical Research Institute of Sweden Our work is concentrated on innovation and the development of value-adding technology. Using Sweden's most extensive and advanced resources for technical evaluation, measurement technology, research and development, we make an important contribution to the competitiveness and sustainable development of industry. Research is carried out in close conjunction with universities and institutes of technology, to the benefit of a customer base of about 9000 organisations, ranging from start-up companies developing new technologies or new ideas to international groups.

SP Technical Research Institute of Sweden SP Report 2012:12 SP Wood Technology ISBN 978-91-87017-28-5 Borås – Skellefteå – Stockholm – Växjö ISSN 0284-5172 Phone: 010-516 50 00

E-mail: [email protected], Internet: www.sp.se

More information on SP publications at www.sp.se/publ