SCREW AND NAIL HOLDING PROPERTIES OF PLYWOOD PANELS REINFORCED WITH GLASS FIBER FABRIC

In this study, plywood and reinforced plywood with glass fiber fabric were produced with rotary peeled poplar veneers using phenol formaldehyde adhesive. Screw and nail holding tests were conducted to obtain the characteristics of the plywood and reinforced plywood boards. One control group and three different test groups were used. The effects of glass fiber fabric on the screw direct withdrawal, screw-head pull-through, and lateral nail resistance were determined. The results of the tests indicated that the reinforced plywood’s resistances to direct withdrawal of the screw, screw-head pullthrough, and lateral nail resistance were significantly higher than those of the unreinforced plywood. In addition, reinforcement of the plywood boards provided longer displacement at maximum load. PROPRIEDADES DE FIXAÇÃO DE PARAFUSOS E PREGOS EM PAINÉIS COMPENSADOS DE MADEIRA REFORÇADOS COM TECIDO DE FIBRA DE VIDRO RESUMO: Neste estudo, painéis compensados e painéis compensados reforçados com tecido de fibra de vidro foram produzidos com lâminas do desenrolamento do álamo usando adesivo de fenol-formaldeído. Testes de fixação de parafusos e pregos foram realizados para obter as características dos painéis compensados e painéis compensados reforçados com tecido de fibra de vidro. Os efeitos do tecido de fibra de vidro sobre o arrancamento direto do parafuso, puxado pela sua cabeça, e resistência lateral do prego foram determinados. Os resultados dos testes indicaram que a resistência do painel reforçado ao arrancamento direto do parafuso, puxado pela sua cabeça, e resistência lateral do prego foram significativamente superiores em relação aos valores encontrados para compensados não reforçados. Além disso, o reforço em painéis compensados resultou em maior elongação na carga máxima.


INTRODUCTION
Solid wood and wood-based composite boards are used in constructing buildings because they have some superior advantages, such as easy processability, low price, good aesthetic appearance, and light weight.But, in addition to those positive features, these materials have some negative features, such as low loadcarrying capacity and other low mechanical properties (DEMIRKIR et al., 2013;GAFF et al., 2015).Different fibers, such as natural fibers and synthetic fibers, are used to reinforce wood-based composite boards (XU et al., 1998a;CAI, 2006;MOHEBBY et al, 2011;GÜNTEKIN and AYDIN, 2015) and load-carrying structural wooden elements (BORRI et al, 2013;BASTERRA et al, 2012).The main aim of these reinforcement related studies was to obtain a new composite material with high mechanical performance.For this, many researchers have investigated reinforced materials using different types of fibers.For example, Wangaard (1964) and Biblis (1965) were the pioneers of reinforcing wood using fiberglass in their experiments.Xu et al. (1998a) investigated the effects of the length of the fibers and their orientation on the elasticity of fiber-reinforced plywood.Xu et al. (1998b) studied the mechanical properties of plywood reinforced by bamboo or jute and reported some positive results.Basterra et al. (2012) studied reinforced duo beams with flax fibers, glass fibers, and carbon fibers.Borri et al. (2013) reinforced timber beams with hemp fibers, flax fibers, basalt fibers, and bamboo fibers.Bal (2014a-b) investigated some of the mechanical properties of reinforced laminated veneer lumber with glass fiber fabric using phenol formaldehyde resin.Bal et al. (2015) investigated some of the technological properties of poplar plywood reinforced with glass fiber fabric and reported some favorable results.
Generally, when wood is used for construction, the covering materials, such as plywood and oriented strand board (OSB), are fastened to the main wooden members, such as solid lumber, laminated veneer lumber, and parallel strand lumber.Different types of screws and nails are used to fasten the members to each other.The durability of construction is dependent on fastening the members at the connection points between the main wooden members and the covering members.In some cases, such as seismic activity and strong winds, failures can occur at the connection points even though the wooden members have very high mechanical performance (HERZOG;YEH, 2006).In these cases, screws and nails may be withdrawn or the cover materials may break away from the head of the fastener.There are limited studies concerned with screw tests of plywood that have been reinforced by various methods.Candan and Akbulut (2014) investigated the flexural properties and resistance to the withdrawal of screws of reinforced plywood using nanomaterials at different loading levels, and some positive results were reported.
In accordance to the previous studies mentioned above, it can be said that the researchers reported some increases of mechanical properties.Generally, previous studies have investigated the flexural properties and performance of the connection points of reinforced materials, such as lumber that has been sawn and structural composite lumbers.There have been no investigations of the screw and nail holding performance of plywood reinforced with glass fibers.Thus, the main aim of this study is to investigate the screw and nail holding performance of plywood reinforced with different combinations of glass fiber fabric.

Material
In this study, plywood panels (control group) were produced with poplar rotary veneers using phenol formaldehyde adhesive.In addition, reinforced plywood panels (test groups) were produced with poplar veneers and glass fiber fabric using the same adhesive.Poplar veneers (Populus deltoides Bartr.I-77/51) were obtained from a plywood factory.The dimension of the veneers were 2.7 x 650 x 2000 mm (thickness x width x length).The moisture content of the veneers was 7±1 %.Phenol formaldehyde (PF) adhesive was obtained from Polisan A.Ş.Some properties of the PF adhesive are described as follows: specific gravity at temperature 20 °C of 1.2 g/cm 3 ; solid content is 47±1%, pH is 10.5-13, viscosity at temperature 20 °C is 250-500 Cp.PF adhesive was used in its pure form.Glass fiber fabric was bought from Fibroteks A.Ş.The weight of the woven glass fiber was 500 g .m -2 , and the knitting type of fabric was 'plainweave'.Glass fiber fabric was obtained as a roll, and was divided into small sheets (600 x 600 mm).

Plywood production
For plywood production, rotary peeled veneers were cut to the dimensions of 2.7 x 600 x 600 mm.Veneers were glued manually using a roller.No additives or fillers were used with the PF adhesive.Adhesive in the amount of 220 g .m -2 was applied on the surfaces of the veneer.Five-ply veneers were used for each panel.The adhesive was applied to the loose side of the veneers.The veneers were piled on each other crosswise.Plywood panels were pressed in a laboratory using an electrically-heated press.The press pressure was 7 kg .cm -2 , the press time was 18 minutes, and the press temperature was 140 °C.The 'A' group of plywood panels was produced using only veneer sheets.The B, C, and D groups of plywood panels were produced with veneers and glass fi ber sheets, as shown in Figure 1 and Figure 2.
Lateral nail resistance was tested on samples with the dimensions of 13 x 50 x 250 mm (thickness x width x length).The lateral nail resistance test was performed as shown in Figure 3C.The plywood test samples were nailed to solid beech wood.No pilot hole was drilled for the nail test samples.The nails were driven into the plywood until the head of the nail was fl ush with the surface of the plywood.The dimensions of the plywood were 13 x 50 x 300 mm, and the dimensions of the main member (beech wood) were 50 x 50 x 200 mm (thickness x width x length).The samples used to test lateral nail resistance were prepared for both the parallel and perpendicular directions to the grain of the surface layers.Twelve test samples were prepared for each group.Common nails that had dimensions of 3 x 60 mm were used in the tests.The nails were hammered manually into the test samples.The nails were driven as nearly perpendicular to the surface of the specimen as possible.The speed of the lateral nail resistance tests was 8 mm .min -1 .

Method
The direct withdrawal of the screw, the screwhead pull-through, and the lateral nail resistance tests were conducted according to ASTM D1761-12 with the exception of the dimensions of the samples.
For the screw direct withdrawal (Figure 3A) and the screw-head pull-through (Figure 3B), the tests were conducted on samples with the dimensions of 13 x 75 x 75 mm (thickness x width x length).The dimensions of the fl athead screw were as follows, total length of 45 mm, length of the threaded portion of 30 mm, diameter of the shank of 4 mm, diameter of the head of 8.5 mm, and diameter of the threaded portion of 4.5 mm.Fourteen test samples were prepared for each group.No pilot holes were drilled for screw-test samples.The test speed was 3 mm .min -1 for the screw direct withdrawal and screw-head pull-through tests.The tests were performed on a Losenhausen testing machine.

RESULTS AND DISCUSSION
Density, maximum load, and displacement at maximum load at the end of tests of the screw direct withdrawal, screw-head pull-through, and lateral nail resistance tests are shown in Table 1.In addition, signifi cance levels of the ANOVA and Tukey test results (shown in lower case) are given in the same table.When the data in Table 1 were analyzed, it was apparent that the density values of the test groups were signifi cantly higher than those of the control group.These results were expected because of the greater density of the glass fi ber.Solid wood and wood-based composite materials have higher strength/density ratios (specifi c strength properties) than other engineering materials, such as concrete, metals, glass, and marble.Many researchers have reported the specifi c mechanical properties of engineered wood products in their studies (TANK; PU, 1997;LEE et al., 1999;BAO et al., 2001;BAL;BEKTAŞ, 2012;BAL;BEKTAŞ, 2014).Therefore, the density increase is the undesirable result of present study.
Table 1 provides the results of the maximum load and displacement at maximum load of the screw direct withdrawal tests.The maximum loads of the B, C, and D groups were higher than for the A group, but the C and D groups had values that were significantly higher than those of the A and B groups (P < 0.001).The increase in the values of the C and D groups was approximately 14% compared to the A group.In addition, displacement at maximum load of the test groups was higher than the control group, but the differences among the groups were not significant (NS).It can be said that the glass fiber fabric and adhesive used to bond glass fiber enhanced the screw withdrawal strength of the C and D groups.
In previous studies, it was determined that many factors affect the screw withdrawal strength of materials, such as the density of the wood, the direction of the grain, the moisture content of the wood, and the type of screws used (WANG et al., 2007;TAJ et al., 2009;ÖZÇIFTÇI, 2009;BAL et al., 2013, ESHAGHI et al., 2013) In the usage areas, the failure of screws occurred due to the threads and due to the failure of the heads of the screws.Therefore, we conducted tests to determine these characteristics of the screws.One of the tests was the screw-head pull-through test.Maximum load and displacement at maximum load of screw-head pull-through test of the B, C, and D groups were greater than those in the control group.The values of the C and D groups were greater than those for the A and B groups, and differences were significant (p < 0.001).The increase of maximum load and displacement at maximum load of the D group were approximately 32 and 50%, respectively.The increase that was obtained was more remarkable than the other test results presented here.Another remarkable result was that the results of the screw-head pull-through tests were greater than those of the tests of the direct withdrawal tests.The reason for this difference was that the diameter of the screw head (8.5 mm) was greater than that of the threads of the screw (4.5 mm).Another reason was the resistance of the glass fiber.Glass fiber is drilled with a diameter of 4.5 mm when the screw is screwed.The screw is pulled from the same hole with a diameter of 4.5 mm when the screw direct withdrawal test is conducted.But, when the screw head is pulled, the head (8.5 mm) of the screw is forced through the 4.5 mm-diameter hole.
Lateral nail resistance is a very important issue for wooden construction and buildings.It can occur at many connection points, but in some connection points at which the sheathing materials (plywood or OSB) are fastened to solid wood or structural composite lumber, the nail's head can be pulled through the sheathing material.This is one of the three major types of failures.The other two types are the nail being withdrawn from the main member and the splitting of the main member.In addition, Rammer (2010) showed that there were some different possibilities of wood bearing and nail bending at the end of lateral nail resistance, such as Modes I-IV.In the present study, Mode III type was observed at the end of the test.
Table 1 provides the lateral nail resistances in the perpendicular and parallel directions.The maximum loads of the B, C, and D groups in both the parallel and perpendicular groups were greater than that of the A group.The differences were significant, as indicated in the bottom line of Table 1.The increases of the perpendicular samples in the C and D groups were approximately 18% greater than the increases in the A group.For the parallel samples, this value was approximately 19%.The most remarkable increase was the displacement at maximum load of the C and D groups for the parallel samples (52%).
The maximum loads of the C and D groups were similar in both directions.The reason for this likely was that four glass fiber sheets were used for the C and D groups in different places.It can be inferred that the placement of the glass fiber sheets had no effect on the lateral resistance if the nails.However, when the related data were analyzed, it was apparent that the maximum load of parallel samples was greater than that of the perpendicular samples.It is well known that solid wood and wood-based composite materials, such as structural composite lumbers, plywood, and OSB, have different properties in the parallel and perpendicular grain directions (POTTER, 1976;ÖZEN, 1981;BIBLIS;CARINO, 2000;BEKHTA et al., 2009;DAOUI et al., 2011;CHANS et al., 2014;BAL;BEKTAŞ, 2014;IWAKIRI et al., 2015;DEMIRKIR;ÇOLAKOĞLU, 2015) because wood is a natural anisotropic material.Related to the plywood, some results were reported by Demirkir and Çolakoğlu (2015); they studied the effect of grain direction on lateral nail strength of structural plywood, and they stated the maximum load of perpendicular samples was signifi cantly greater than that of parallel samples.In the same study, they noted that Potter (1976) determined that the lateral nail resistance was greater for the perpendicular samples than for parallel samples, while Hunt and Bryant (1984) determined opposite results, and Pirvu (2008) reported that the grain direction had no effect on the lateral nail strength of plywood.In previous studies of these discrepancies, some variables may have greater effects, such as the ply number of the plywood and the thickness of the veneer.It is known that as the number of ply increases in plywood, the discrepancy between parallel and perpendicular samples decreases (ÖZEN, 1981).But, in the present study, there was the effect of the glass fi bers on the lateral nail resistance.The difference between the A group's perpendicular and parallel samples differed from the C and D groups.In a previous study it was reported that the support of the glass fi ber fabric decreased the inequality in the modulus of elasticity and the modulus of rupture tests between the parallel and perpendicular samples (BAL et al., 2015).At the end of tests, the C and D groups provided signifi cant improvements.Therefore, the discussion was focused on the comparison of the A and D groups.The load-deformation curves of screw-head pull-through test samples of the A and D groups are given in Figure 4.The differences between the two groups are evident.The load-deformation curves of the D group were wavy after the elastic zone.It can be said that the effective factor for fl uctuation was the glass fi ber fabric.Figure 5 shows the load-deformation curves of the screw direct withdrawal test samples of groups A and D. In Figure 5, the fl uctuation is lower than in Figure 4.This result was expected due to the dimensions of the screw head.Figure 6 shows the load-deformation curves of the lateral nail resistance test of the perpendicular samples, and Figure 7 shows the load-deformation curves of the lateral nail resistance test of the parallel samples.Another result that can be inferred from the fi gures is that the plastic region of test groups (B, C, D) was greater than that of the control group.
the data that were obtained, we reached the following conclusions.The maximum load of the C and D groups was signifi cantly greater than that of control groups for all of the tests.The differences between the C and D groups were not signifi cant.The maximum loads of lateral nail resistance of the parallel groups were greater than those of the perpendicular groups.Displacements at maximum load of screw direct withdrawal and lateral nail resistance in the perpendicular samples were insignifi cant compared to the control samples.The property tested here that showed the greatest increase in the reinforced plywood panels was the screw head pull-through in both the maximum load and displacement at maximum load.
Consequently, it can be said that the effects of reinforcement with glass fi ber fabric on the screw and nail strength of plywood are not very high.But, this fi nding should be evaluated and compared with the results of previous studies of reinforced plywood.In future experiments, the screw or nail strength and the strength of the reinforced connection points should be investigated completely.

CONCLUSIONS
In this study, the effects of glass fi ber fabric on the screw direct withdrawal, screw-head pull-through, and lateral nail resistance were determined.According to

FIGURE 1
FIGURE 1 Schematic representation of the plywood groups.

FIGURE 2
FIGURE 2Poplar veneers and glass fi ber sheets for B group.

FIGURE 6 FIGURE 7
FIGURE 6 Load-deformation curves of lateral nail resistance test of perpendicular samples

TABLE 1
Screw and Nail test data and ANOVA and Tukey results SDW: screw direct withdrawal, SHPT: screw-head pull-through, LNR: lateral nail resistance, D: density, Lmax: maximum load, Dmax: displacement at maximum load.x: arithmetic mean, ss: standard error, cov: coefficient of variation.