Wood Fibre Insulation Boards As Load-Carrying Sheathing Material of Wall Panels

First published in:

Wood Fibre Insulation Boards as Load-Carrying Sheathing Material of Wall Panels

Gunnar Gebhardt1, Hans Joachim Blaß2

ABSTRACT: To date wood fibre insulation boards (WFIB) are used as thermal and acoustic insulation in timber construction. As wood-based panels, WFIB are also suited for the transfer of loads caused by wind and earthquakes in timber frame construction, where they may replace plywood, particle boards or OSB. This new field of application of WFIB has been analysed in a research project at Karlsruhe Institute of Technology. In order to calculate the load-carrying capacity of wall and roof panels the characteristics of WFIB as well as those of joints with WFIB and dowel-type fasteners need to be known. In tests these characteristics were determined. A proposal is made for the calculation of timber-WFIB-joints with dowel type fasteners. Furthermore the stiffness properties of timber-WFIB-joints were determined. In this paper the results of the tests are presented.

KEYWORDS: Wood fibre insulation board, wall panel, sheathing material

embedding strength of nails in WFIB and the crown INTRODUCTION AND pull-through resistance of wide-crown staples in WFIB. CHARACTERISTICS OF WFIB 12 The test results verified the calculation model and the Wood-based sheathing materials in timber frame stiffness properties of timber-WFIB-joints were construction normally are plywood, particle boards or evaluated. Furthermore the shear strength and the shear OSB. A further possibility is sheathing made of gypsum modulus of WFIB were determined. plasterboards. These panels have a relatively high The raw material of WFIB is wood chips. WFIB are density caused by the compression and adhesion of the fabricated in two different manufacturing processes. In different raw materials. The heat insulation properties do both the wood chips are first thermo-mechanically not equal those of typical insulation materials like pulped. In the wet process the wood fibres are mixed mineral or wood fibre. For the heat insulation of outer with water and further aggregate into a suspension. walls therefore additional insulation material is required. Afterwards the boards are formed and dried. For the Due to the increasing demands regarding energy and bonding of the wood fibres only the wood´s own CO2 saving the insulation thickness increases. Wood cohesiveness (predominantly lignin resin) is used. Due to fibre insulation materials produced as panels are the high use of energy for the drying process, boards therefore suited for the use in shear walls in timber frame with higher thicknesses are manufactured by gluing raw construction if they are able to carry the shear loads in single-layer boards to multilayer boards, where also the panel itself and if the connections between WFIB different density boards may be combined. In the dry and timber members have a sufficient strength and process the wood fibres are dried and sprayed with resin. stiffness. This new field of application of WFIB was Afterwards the boards are formed and the resin hardens. analysed in a research project at Karlsruhe Institute of In this process boards with thicknesses up to 240 mm are Technology. In this paper a proposal is given to calculate manufactured as single-ply boards. WFIB are used in the load-carrying capacity of timber-WFIB-joints with different parts of buildings. In roofs, rain-tight sub plates dowel-type fasteners. For this purpose the load-carrying are fixed as sub decking. A further insulation is possible capacity of joints in timber and WFIB is determined as above-rafter or interim-rafter insulation with universal according to Johansen’s yield theory and an extension of insulation boards. In walls, plaster baseboards may be this theory. Tests were carried out to evaluate the used in composite thermal insulation systems. 11 different WFIB of three different manufacturers were 1 Gunnar Gebhardt, Timber structures and building selected to analyse the characteristics needed for the use construction, Karlsruhe Institute of Technology, R.- as sheathing in shear walls. The nominal density range Baumeister-Platz 1, 76131 Karlsruhe, Germany. Email: was between 110 kg/m3 and 270 kg/m3. The densities [email protected] 2 and moisture contents of the tested boards were Hans Joachim Blaß, Timber structures and building determined. The measured moisture content was 7.3% to construction, Karlsruhe Institute of Technology, R.- Baumeister-Platz 1, 76131 Karlsruhe, Germany. Email: 10.3%. [email protected]

EVA-STAR (Elektronisches Volltextarchiv – Scientific Articles Repository) http://digbib.ubka.uni-karlsruhe.de/volltexte/1000018785

ANALYSIS OF WALL PANELS the extended calculation model. The yield moments of In analysis methods the load-carrying capacity of the the fasteners were evaluated according to EN 409. The connection between timber member and sheathing panel embedding strength of staples in WFIB was calculated is required. The analysis according to German design assuming the validity for diameters smaller than the code DIN 1052 in addition requires the shear strength of tested ones. The embedding strength of nails in WFIB the sheathing panel. The connection’s load-carrying was determined in tests and used for the calculation. The capacity is calculated according to Johansen’s yield embedding strength of the fastener in the timber member theory considering the different failure mechanisms. was calculated according to DIN 1052 considering the In joints with fasteners driven in through a counter specimen densities. The withdrawal resistance and the batten the inclination of the fastener in failure modes 1a pull-through resistance were calculated according to DIN and 2b involves embedding resistances in the counter 1052 for nails and staples. The crown pull-through batten. Hence the related equations for these failure resistance for wide-crown staples was determined in mechanisms are extended. The load path is from the stud tests. into the sheathing panel. There is no resulting load in the The tests were carried out according to EN 26891. The counter batten. Force and moment equilibrium, maximum load was reached at a displacement of 15 mm. considering the existing embedding strength distribution Although failure mode 1b was governing in the and the yield moment, result in equations for the calculation for nails, failure mode 3 was observed in the extended failure mode 1a and for the extended failure tests. This may be explained by friction between the joint mode 2b. members. The relative deformations between WFIB and The load-carrying capacity increases with increasing stud were measured at each of the four fasteners. The axial load-carrying capacity of the fastener. If the analysis of the joint stiffness considers a linear load- fastener is driven in directly into the sheathing panel the displacement behaviour up to 40% of the maximum axial load-carrying capacity depends on the withdrawal load. Due to a non-linear load-displacement behaviour parameter of the fastener in the timber member and the the analysis of the stiffness was calculated for a constant head/crown pull-through resistance in the sheathing displacement of 0.3 mm. By means of a multiple panel or a counter batten, if the fastener is driven in regression analysis an equation was derived to calculate through a counter batten the stiffness of timber-WFIB-joints. BASIC PARAMETERS OF DOWEL-TYPE SHEAR STRENGTH AND SHEAR FASTENERS IN WFIB MODULUS OF WFIB About 600 embedment tests with nails were performed The shear strength and the shear modulus of WFIB were according to EN 383 in order to evaluate the embedding determined according to EN 789. The WFIB specimens strength of dowel-type fasteners in WFIB. The test were reinforced by four lateral timber boards connected results for the embedding strength correlate with the with the WFIB test specimen by five pre-stressed bolts. measured densities of the test specimens and the fastener The correlation between shear strength, density and diameter. By means of a multiple regression analysis an board thickness was analysed. The correlation equation was derived for the embedding strength of coefficient between shear strength and board thickness is WFIB. In order to calculate the rope effect the crown r = -0.544, while the correlation coefficient between pull-through resistance of wide-crown staples in WFIB shear strength and density is r = 0.847 and between was examined. According to EN 1383 about 100 tests density and board thickness is r = -0.429. with RTSP and PB were carried out. The displacement at Furthermore the correlation coefficient between shear maximum load correlates with the thickness of the modulus and density is r = 0.894. WFIB. By means of a multiple regression analysis an CONCLUSIONS equation was derived to estimate the crown pull-through To date wood fibre insulation boards are used as acoustic resistance of staples in WFIB. and thermal insulation. As wood-based panels they can LOAD-CARRYING CAPACITY AND also be used for the load transfer in wall panels. The load STIFFNESS OF TIMBER-WFIB-JOINTS transfer requires a sufficient load-carrying capacity of In order to verify the results obtained in tests and to the joint between studs and sheathing and sufficient examine the stiffness of joints, further tests were shear strength of the WFIB. In order to calculate the load performed. In tests with joints the calculated values carrying capacity of joints between timber and WFIB the considering the determined parameters are checked. embedding strength of nails in WFIB and the crown Furthermore the stiffness of timber-WFIB-joints was pull-through resistance of wide-crown staples in WFIB evaluated. Rain-tight sub plates of three manufacturers in were determined. Johansen’s yield theory was extended three thicknesses, respectively, and plaster baseboards to consider the influence of a counter batten on the load were used as sheathing panel. Nails and staples were carrying capacity. The stiffness of joints between timber driven in through a counter batten. Wide-crown staples and WFIB was evaluated in further tests. The shear were driven in directly. strength and shear modulus of WFIB were also The load-carrying capacity may be obtained according to determined in tests. The results serve as a basis for the Johansen’s yield theory considering the test results and analysis of wood fibre insulation boards as sheathing panel in timber frame construction.

Wood Fibre Insulation Boards as Load-Carrying Sheathing Material of Wall Panels

Gunnar Gebhardt1, Hans Joachim Blaß2

In order to calculate the load-carrying capacity of wall and roof panels the characteristics of WFIB as well as those of joints with WFIB and dowel-type fasteners need to be known. In tests these characteristics were determined. A proposal is made for the calculation of timber-WFIB-joints with dowel type fasteners. Furthermore the stiffness properties of timber-WFIB-joints were determined. In this paper the results of the tests are presented.

KEYWORDS: Wood fibre insulation board, wall panel, sheathing material

1 INTRODUCTION 12 Johansen’s yield theory and an extension of this theory. Tests were carried out to evaluate the embedding Wood-based sheathing materials in timber frame strength of nails in WFIB and the crown pull-through construction normally are plywood, particle boards or resistance of wide-crown staples in WFIB. The test OSB. A further possibility is sheathing made of gypsum results verified the calculation model and the stiffness plasterboards. These panels have a relatively high properties of timber-WFIB-joints were evaluated. density caused by the compression and adhesion of the Furthermore the shear strength and the shear modulus of different raw materials. The heat insulation properties do WFIB were determined. not equal those of typical insulation materials like mineral or wood fibre. For the heat insulation of outer 2 CHARACTERISTICS OF WFIB walls therefore additional insulation material is required. Due to the increasing demands regarding energy and The raw material of WFIB is wood chips. WFIB are CO2 saving the insulation thickness increases. Wood fabricated in two different manufacturing processes. In fibre insulation materials produced as panels are both the wood chips are first thermo-mechanically therefore suited for the use in shear walls in timber frame pulped. construction if they are able to carry the shear loads in In the wet process the wood fibres are mixed with water the panel itself and if the connections between WFIB and further aggregate into a suspension. Afterwards the and timber members have a sufficient strength and boards are formed and dried. For the bonding of the stiffness. This new field of application of WFIB was wood fibres only the wood’s own cohesiveness analysed in a research project at Karlsruhe Institute of (predominantly lignin resin) is used. Due to the high use Technology. of energy for the drying process, boards with higher In this paper a proposal is given to calculate the load- thicknesses are manufactured by gluing raw single-layer carrying capacity of timber-WFIB-joints with dowel type boards to multilayer boards (figure 2), where also fasteners. For this purpose the load-carrying capacity of different density boards may be combined. joints in timber and WFIB is determined according to In the dry process the wood fibres are dried and sprayed with resin. Afterwards the boards are formed and the 1 Gunnar Gebhardt, Timber structures and building resin hardens. In this process boards with thicknesses up construction, Karlsruhe Institute of Technology, R.- to 240 mm are manufactured as single-ply boards (see Baumeister-Platz 1, 76131 Karlsruhe, Germany. Email: figure 1). [email protected] 2 Hans Joachim Blaß, Timber structures and building construction, Karlsruhe Institute of Technology, R.- Baumeister-Platz 1, 76131 Karlsruhe, Germany. Email: [email protected]

wall – plaster baseboard and insulation Figure 3 (cont.): Applications of WFIB

11 different WFIB of three different manufacturers were selected to analyse the characteristics needed for the use as sheathing in shear walls. The nominal density range was between 110 kg/m3 and 270 kg/m3. The densities and moisture contents of the tested boards were Figure 1: WFIB – single-ply boards determined. The measured moisture content was 7.3% to 10.3%. Characteristic densities for the different types of WFIB are given in table 1.

Table 1: Proposed characteristic density of WFIB

Characteristic WFIB type density in kg/m3 Rain-tight sub plate (RTSB) 200 Plaster baseboard (PB) 150 150 Universal insulation board (UIB) 100

3 ANALYSIS OF WALL PANELS 3.1 ANALYSIS ACCORDING TO GERMAN Figure 2: WFIB – multilayer boards DESIGN CODE DIN 1052:2008-12 WFIB are used in different parts of buildings. In roofs, The analysis according to German design code DIN rain-tight sub plates (RTSP) are fixed as sub decking. A 1052 [1] is a simplified analysis for wall and roof panels. further insulation is possible as above-rafter or interim- Three different failure modes are considered. Beside the rafter insulation with universal insulation boards (UIB). failure of the connection between timber and sheathing In walls, plaster baseboards (PB) may be used in material, shear failure and buckling failure of the composite thermal insulation systems. In figure 3 the sheathing panel are considered. The shear flow per wall different applications are shown. panel length is obtained using Equation (1): F s = vd, (1) vd,0, where Fv,d = design value of horizontal load acting on wall

roof – rain-tight sub plate (RTSP) panel and = length of wall panel.

The shear strength per wall panel length is calculated according to Equation (2):

⎧ Rd ⎫ roof – rain-tight sub plate (RTSP) and interim rafter insulation ⎪ kv1 ⋅ ⎪ av ⎪ ⎪ fkkftvd,0,=⋅⋅⋅min ⎨ v 1 v 2 vd , ⎬ (2) ⎪ 2 ⎪ ⎪ t ⎪ kkvv12⋅⋅ f vd , ⋅⋅35 ⎪⎩⎭ar ⎪ where kv1 = factor considering the joints at the panel boundaries (kv1 = 1, if panel boundaries fixed at all sides, else kv1 = 0.66), roof – rain-tight sub plate (RTSP) and above rafter insulation kv2 = factor considering if the sheathing is one-sided Figure 3: Applications of WFIB (kv2 = 0.33) or both-sided (kv2 = 0.5), Rd = design value of load-carrying capacity of the involves embedding resistances in the counter batten. joint between sheathing and timber frame, Hence the related equations for these failure mechanisms av = spacing of fasteners, are extended. fv,d = design value of shear strength of sheathing panel, 4 EMBEDDING STRENGTH OF WFIB t = thickness of sheathing panel and Tests with nails were performed according to EN 383 [3] ar = spacing of vertical studs. to evaluate the embedding strength of mechanical 3.2 ANALYSIS ACCORDING TO EN 1995-1-1 fasteners in WFIB. The test set-up is shown in figure 4. In preliminary tests an influence of the angle between The analysis according to EN 1995-1-1 [2] is a force and production direction could not be identified. simplified analysis for wall and roof panels similar to the About 600 embedment tests with five different diameters analysis in DIN 1052. However, the analysis only (d = 3.1/3.4/3.8/4.6/5.0 mm) were carried out. The test considers the failure of the connection between timber results for the embedding strength correlate with the and sheathing material. The load-carrying capacity of a measured densities of the test specimens. In figure 5 the wall panel is obtained using Equation (3): embedding strength is plotted vs. the density. The F ⋅bc⋅ correlation coefficient is r = 0.880. F = f ,Rd i i (3) ivRd,, s where Ff,Rd = design value of the load-carrying capacity of a single fastener, bi = length of wall panel, s = spacing of fasteners.

Furthermore, the analysis considers the geometry of the wall panel by a factor ci, see Equation (4):

⎧ 1 for b10≥ b ⎪ ci = ⎨b (4) i for b< b ⎪ 10 ⎩b0 where b0 = h/2, h = height of wall panel.

3.3 NECESSARY TESTS AND ANALYSIS In both analysis methods the load-carrying capacity of the connection between timber member and sheathing Figure 4: Test - embedding strength of nails in WFIB panel is required. The analysis according to DIN 1052:2008-12 in addition requires the shear strength of There is a less pronounced correlation between the the sheathing panel. The connection’s load-carrying embedding strength and the thickness of the boards. A capacity is calculated according to Johansen’s yield negative correlation exists between the embedding theory in both standards and depends on the geometry of strength and the fastener diameter. Due to a correlation the connection (thicknesses of the jointed members and between the density and the thickness and the higher diameter of the fastener), the yield moment of the correlation between embedding strength and thickness, fastener and the embedding strengths of the jointed the parameters diameter and density are considered in a members. The load-carrying capacity may be increased regression model for the embedding strength. By means if the fastener can be loaded axially apart from loading of a multiple regression analysis, equation (5) was laterally. If the fastener is driven in directly into the derived for the embedding strength fh of WFIB. sheathing panel the axial load-carrying capacity depends on the withdrawal parameter of the fastener in the timber −−5 2.04 0.74 2 fdh =⋅⋅⋅18.5 10 ρ in N/mm (5) member and the head/crown pull-through resistance in the sheathing panel or a counter batten, if the fastener is where driven in through a counter batten. ρ = density of the WFIB in kg/m3 and Johansen’s theory considers six different failure d = diameter of the fastener in mm. mechanisms for single shear joints. In the first three failure mechanisms the embedding strength of one or In figure 6 the embedding strength values from the tests both of the jointed members is reached. In failure modes are plotted vs. the calculated embedding strength values. four and five in addition one plastic hinge appears. In The correlation coefficient is r = 0.916. The slope of the failure mode six two plastic hinges are formed. In joints regression straight is m = 1.02 and the y-intercept is with fasteners driven in through a counter batten the b = -0.090. inclination of the fastener in failure modes 1a and 2b WFIB was examined. The test set-up is shown in 11 figure 7.

2 10 y = 0.0368x - 3.90 9 r = 0.880 8 n = 608 7 6 5 4 3 RTSP 2 PB Embedding strength in N/mm strength Embedding 1 UIB 0 0 50 100 150 200 250 300 350 400 3 Density in kg/m Figure 5: Embedding strength vs. density

2 11 y = 1.02x - 0.090 10 r = 0.916 Figure 7: Test – Crown pull-through resistance of wide- 9 n = 608 crown staples in WFIB 8 7 According to EN 1383 [4] about 100 tests with RTSP 6 and PB were carried out. For each of the 15 different 5 WFIB at least four tests were performed. The 4 3 displacement at maximum load correlates with the RTSP 2 thickness of the WFIB. In figure 8 the displacement at PB 1 UIB maximum load is plotted vs. the thickness of the WFIB. Tested embedding strength in N/mm strength embedding Tested 0 01234567891011 40 y = 0.500x + 0.115 Calculated embedding strength in N/mm2 35 r = 0.893 n = 101 Figure 6: Embedding strength from tests vs. calculated 30 embedding strength 25 20 Characteristic values of the embedding strength may be 15 calculated considering the proposed characteristic 10 densities for the different types of WFIB (table 2). RTSP 5 PB Table 2: Characteristic embedding strength in N/mm2 Displacement at maximum load in mm 0 0 10203040506070 Characteristic Thickness in mm Characteristic WFIB embedding strength density in kg/m3 Figure 8: Displacement at maximum load vs. thickness in N/mm2 −0.75 By means of a multiple regression analysis, equation (6) RTSP 200 fdh,k =⋅8.88 was derived to estimate the crown pull-through PB 150 fd=⋅4.25 −0.75 h,k resistance Rax,2 of staples in WFIB. −0.75 UIB 150 fdh,k =⋅3.53 1.17 0.95 −0.75 Rtax,2 = 0.040⋅⋅ρ in N (6) UIB 100 fdh,k =⋅1.57 where ρ = density of the WFIB in kg/m3 and 5 CROWN PULL-THROUGH t = thickness of the WFIB in mm. RESISTANCE OF WIDE-CROWN STAPLES IN WFIB In figure 9 the crown pull-through resistance values from tests are plotted vs. the calculated values. The correlation The load-carrying capacity according to Johansen’s yield coefficient is r = 0.814. The slope of the regression theory depends on the geometry of the connection straight is m = 1.01 and the y-intercept is b = -0.01. The (thicknesses of the connected members and diameter of test results of a RTSP with relatively high thickness and the fastener), the embedding strengths of the members density cannot be explained by the regression model, the (for WFIB presented in chapter 4) and the yield moment test values are higher than the calculated values. Here, of the fastener. The load-carrying capacity increases with further tests are needed. increasing axial load-carrying capacity of the fastener. The rope effect depends on the withdrawal strength of the fastener in the first member and the head/crown pull- through resistance in the second member. Hence the crown pull-through resistance of wide-crown staples in 2,5 y = 1.01x - 0.01 ⎡ 412ββ()+ M ⎤ ⎢ 2 22y ⎥ r = 0.814 21ββ22()+ ++2 2,0 n = 97 ⎢ fdth12⋅⋅ ⎥ ftdh,1⋅⋅ 2 ⎢ 2 ⎥ R = ⎛⎞t (9) 12+ β ⎢ 3 ⎥ 1,5 2 ββ23⋅+()12 β 2⎜⎟ ⎢ t ⎥ ⎢ ⎝⎠2 ⎥ ⎢ ⎥ 1,0 ⎣−β2 ⎦ where 0,5 RTSP t3 = thickness of the counter batten, PB

Tested crown pull-through resistance in in kN resistance crown pull-through Tested f = embedding strength of the counter batten and 0,0 h,3 0,00 0,25 0,50 0,75 1,00 1,25 1,50 f β = h,2 Calculated crown pull-through resistance in kN 2 fh,1 Figure 9: Crown pull-through resistance from tests vs. f calculated crown pull-through resistance h,3 β3 = fh,1 Characteristic values of the crown pull-through resistance may be calculated considering the proposed characteristic densities for the different types of WFIB F (equation (7)).

1.17 0.95 Rtax,2,k=⋅⋅0.032 ρ k in N (7) where 3 ρk = density of the WFIB in kg/m .

6 CALCULATION OF JOINTS BETWEEN TIMBER AND SHEATHING PANEL The load-carrying capacity may be calculated according f f f to Johansen’s yield theory considering the different h,1 h,2 h,3 failure mechanisms. If the fastener is driven in through a counter batten, in failure modes 1a and 2b an extension of the existing equations is necessary. In failure modes F 1a and 2b the embedding strength in the counter batten is reached and this effect increases the load-carrying Figure 10: Consideration of the counter batten in failure capacity. The load path is from the stud into the mode 1a sheathing board. There is no resulting load in the counter batten. Force and moment equilibrium, considering the F existing embedding strength distribution and the yield moment, result in equation (8) for the extended failure mode 1a and equation (9) for the extended failure mode 2b. The failure modes with the influence of the counter batten are shown in figure 10 and figure 11.

M ⎡⎤2 ⎡⎤⎛⎞⎛⎞tt ⎢⎥ββ++++212 ⎢⎥22 ⎢⎥22⎜⎟⎜⎟ ⎢⎥⎝⎠⎝⎠tt11 ⎢⎥⎣⎦ ftd⋅⋅ ⎢⎥22 h,1 1 ⎛⎞t ⎛⎞t R = ⎢⎥3 2 3 ββββ2232⎜⎟+⋅()1 + ⎜⎟ (8) 1+ β2 ⎢⎥ ⎝⎠tt11 ⎝⎠ f f f ⎢⎥ h,1 h,2 h,3 ⎢⎥⎛⎞t2 ⎢⎥−+β2 ⎜⎟1 t ⎣⎦⎝⎠1 F

Figure 11: Consideration of the counter batten in failure mode 2b

120 t F 7 LOAD-CARRYING CAPACITY AND 80 STIFFNESS OF TIMBER-WFIB- JOINTS WITH NAILS AND STAPLES 200 200 In order to verify the results obtained in the tests 60

presented in chapter 4 and 5 and to examine the stiffness 60 of joints, further tests were performed. In the previous paragraphs the basic parameters 400 (embedding strength and crown pull-through resistance) 160 for the calculation of the load-carrying capacity of timber-WFIB-joints were determined. The calculation method according to Johansen’s yield theory was 200 extended for fasteners driven in through a counter batten. In tests with joints the calculated values considering the determined parameters are checked. Furthermore the F 100 100 stiffness of timber-WFIB-joints was evaluated. RTSP of three manufacturers in three thicknesses, respectively, and PB were used as sheathing panel. Nails and staples Figure 13: Test specimen for test of timber-WFIB-joints were driven in through a counter batten. Wide-crown with wide-crown staples driven in directly staples were driven in directly. In figure 12 and figure 13 the test specimens for the tests with nails and staples are shown. In figure 14 the test set-up is shown.

The load-carrying capacity may be obtained according to Johansen’s yield theory considering the test results and the extended calculation model. The yield moments of the fasteners were evaluated according to EN 409 [5]. The embedding strength of staples in WFIB was calculated according to equation (1) assuming the validity for diameters smaller than the tested ones. The embedding strength of nails in WFIB was determined in tests and used for the calculation. The embedding strength of the fastener in the timber member was calculated according to DIN 1052 [1] considering the specimen densities. The withdrawal resistance and the pull-through resistance were calculated according to DIN 1052 [1] for nails and staples. The crown pull-through resistance for wide-crown staples was determined in tests. Figure 14: Test specimen for test of timber-WFIB-joints 120 t 30/ 40 F 80 The tests were carried out according to EN 26891 [6]. 50 The maximum load was reached at a displacement of 15 mm. Although failure mode 1b was governing in the

80 calculation for nails, failure mode 3 was observed in the

200 tests. This may be explained by friction between the joint 60 members. The relative deformations between WFIB and 60 stud were measured at each of the four fasteners. The analysis of the joint stiffness considers a linear load- 400 displacement behaviour up to 40% of the maximum load. In the tests non-linear load-displacement behaviour was observed. Due to this the analysis of the stiffness was calculated for a constant displacement of 0.3 mm. The load-carrying capacity from tests is plotted vs. the calculated load-carrying capacity in figure 15 for nails, F in figure 16 for staples and in figure 17 for wide-crown 100 100 staples.

Figure 12: Test specimen for test of timber-WFIB-joints with nails and staples driven in through a counter batten 2000 1800 2200 k = 1.00 · k - 0.404 1600 2000 s,test s,cal.

1400 1800 r = 0.898 n = 90 1200 1600 1000 1400 1200 800 1000 600 n = 27 800 400 Staple 12 mm d = 3.8 600 200 Staple 27 mm Tested stiffness in N/mm Tested stiffness

Tested load-carrying capacity in N 400 d = 4.6 Nail 3.8 mm 0 200 Nail 4.6 mm 0 200 400 600 800 1000 1200 1400 1600 1800 2000 0 Calculated load-carrying capacity in N 0 200 400 600 800 1000 1200 1400 1600 1800 2000 2200 Calculated stiffness in N/mm Figure 15: Load-carrying capacity from tests vs. calculated load-carrying capacity for nails driven in Figure 18: Stiffness from tests vs. calculated stiffness through a counter batten

8 SHEAR STRENGTH AND SHEAR 2200 2000 MODULUS OF WFIB 1800 1600 The shear strength and the shear modulus of WFIB were 1400 determined according to EN 789 [7]. The geometry of 1200 the test specimens and the test set-up are shown in figure 1000 19 and figure 20. The WFIB specimens were reinforced 800 n = 27 by four lateral timber boards connected with the WFIB 600 t = 18 mm test specimen by five pre-stressed bolts. 400 t = 22 mm 200 Tested load-carrying capacity in N t = 35 mm 0 440 0 200 400 600 800 1000 1200 1400 1600 1800 2000 2200 145 150 145 F Calculated load-carrying capacity in N R = 150 mm 50 Figure 16: Load-carrying capacity from tests vs. 150 calculated load-carrying capacity for staples driven in through a counter batten 200

50 500 900 1200 50

1000 25 5050 25 200

800 150 600 n = 36 R = 150 mm 50

400 RTSP, t = 18 mm RTSP, t = 22 mm 200 RTSP, t = 35 mm Figure 19: Geometry of test specimen and test set-up

Tested load-carrying capacity in N Tested PB 0 0 200 400 600 800 1000 1200 Calculated load-carrying capacity in N Figure 17: Load-carrying capacity from tests vs. calculated load-carrying capacity for wide-crown staples

By means of a multiple regression analysis, equation (10) was derived to calculate the stiffness of a timber- WFIB-joint. In figure 18 the stiffness from tests is plotted vs. the calculated stiffness. The correlation coefficient is r = 0.898.

0.80 0.30− 0.32 1.29 Kser=⋅1.25 ρρ WFIB ⋅⋅⋅td in N/mm (10) where 3 ρWFIB = density of the WFIB in kg/m , ρ = density of timber in kg/m3, t = thickness of WFIB in mm and d = diameter of fastener in mm. Figure 20: Test specimen during test The correlation between shear strength, density and carrying capacity of joints between timber and WFIB the panel thickness was analysed. The correlation coefficient embedding strength of nails in WFIB and the crown between shear strength and board thickness is r = -0.544, pull-through resistance of wide-crown staples in WFIB while the correlation coefficient between shear strength were determined. Johansen’s theory was extended to and density is r = 0.847 and between density and board consider the influence of a counter batten on the load thickness is r = -0.429. The shear strength may be carrying capacity. The stiffness of joints between timber assessed by Equation (11). Figure 21 shows the shear and WFIB was evaluated in further tests. The shear strength vs. the density. strength and shear modulus of WFIB were also determined in tests. The results serve as a basis for the

−62.39 2 analysis of wood fibre insulation boards as sheathing fv =⋅⋅1.30 10 ρ in N/mm (11) panel in timber frame construction. where ρ = density of the WFIB in kg/m3. REFERENCES

[1] DIN 1052:2008-12 Design of timber structures – 1,6 General rules and rules for buildings 2 r = 0.847 1,4 n = 291 [2] EN 1995-1-1 Eurocode 5: Design of timber 1,2 structures – Part 1-1: General – Common rules and 1,0 rules for buildings [3] EN 383:1993-10 Timber structures; Test methods; 0,8 Determination of embedding strength and 0,6 foundation values for dowel type fasteners RTSP 0,4 [4] EN 1383:2000-03 Timber structures – Test methods PB 0,2 UIB – Pull through resistance of timber fasteners Shear strength from tests in N/mm from tests Shear strength 0,0 [5] EN 409:1993-10 Timber structures; Test methods; 0 100 200 300 400 Determination of the yield moment of dowel type 3 Density in kg/m fasteners; Nails Figure 21: Shear strength from tests vs. density [6] EN 26891:1991-07 Timber structures; Joints made with mechanical fasteners; General principles for the The shear modulus is plotted vs. the density in Figure 22. determination of strength and deformation The correlation coefficient between shear modulus and characteristics (ISO 6891:1983) density is r = 0.894. The shear strength may be assessed [7] EN 789:2005-01 Timber structures – Test methods – by Equation (12). Determination of mechanical properties of wood- based panels G =⋅⋅9.03 10−42.13ρ in N/mm2 (12) where ρ = density of the WFIB in kg/m3.

250 2 r = 0.894 n = 208 200

150

100 RTSP 50 PB UIB

Shear modulus from tests in N/mm from modulus Shear 0 0 100 200 300 400 Density in kg/m3

Figure 22: Shear modulus from tests vs. density

9 CONCLUSIONS To date wood fibre insulation boards are used as acoustic and thermal insulation. As wood-based panels they can also be used for the load transfer in wall panels. The load transfer requires a sufficient load-carrying capacity of the joint between studs and sheathing and sufficient shear strength of the WFIB. In order to calculate the load