Mechanical Properties of Finger-Jointed Wood from Composite Utility Poles Made of Small Diameter Timber

Engineering small diameter timber into structural members may provide an effi cient way to utilize low-value material obtained after forest thinning operations. This study evaluates the strength and stiffness properties of fi nger-jointed and solid wood small clear samples cut from composite poles made of small diameter timber. The strength and stiffness of fi nger-jointed small clear samples were compared with the strength and stiffness of solid wood small clear samples and the strength and stiffness of composite poles. Finger-jointed samples tested in a perpendicular orientation, yielded the lowest bending strength but were not signifi cantly lower than samples tested in a parallel orientation. Therefore, fi nger joint orientation was not a signifi cant factor regarding the strength of the poles. The bending strength of composite poles was usually lower than the strength of the solid wood samples but higher than the strength of fi nger-jointed samples cut from the poles. However, the bending stiffness of the composite poles was substantially higher than the bending stiffness of both solid wood and fi nger-jointed samples cut from the poles.


INTRODUCTION 1. UVOD
Wildfi res and forest diseases and insects are major threats to a forest.Studies have shown that silvicultural techniques such as thinning offer the most promising and long-lasting means of preventing insect attacks (Nebeker et al., 1985).Catastrophic wildfi res have also created an incentive to manage forest fuel loading and restore healthy forests via the removal of bio-fuels by thinning operations.The restoration to a healthy forest and the minimization of forest wildfi res both require thinning of overstocked forests.As a result, an ample supply of small-diameter timber (SDT) will continue for the foreseeable future.
The current processing options for using SDT for value-added products, however, are limited.Most SDT produced in the southeastern U.S. is used as feedstock for the production of oriented strand board or pulp and paper (Hodges et al., 2005).According to Wolfe (2000), the value of roundwood from SDT can be twice that in the square form, and nine times that of wood chips.Round timber has less strength variability and more mature wood in the surface rings.Therefore, engineering SDT into structural members may greatly improve the value of SDT and its utilization effi ciency.
In a previous report, we have evaluated a novel technique for making laminated utility poles using SDT (Piao et al., 2015).Eight laminated utility poles, each consisting of four tapered round logs, were fabricated.Each round log was fi nger-jointed together by round segments made from southern pine (Pinus spp.)SDT.The eight laminated utility poles along with three commercial solid wood poles were mechanically tested in a cantilever mode.In this report, fi nger joint samples and solid wood samples were cut from seven of the eight laminated utility poles.The mechanical properties of roundwood fi nger joints were not found in the literature.The objective of this study was to determine the strength and stiffness of fi nger joints along each pole and compare these values with the strength and stiffness of solid wood next to the fi nger joints in the poles.

MATERIJAL I METODE
The fabrication details of laminated utility poles is described in Piao et al., (2014) and is briefl y described below.A total of three hundred and fi fty-six small-diameter, southern pine (Pinus spp.) logs were collected for an on-going series of studies.Logs were collected during thinning operations conducted in typical southern pine plantation forests.The butt diameters of the logs ranged from 3.6 to 12.8 cm.All logs were approximately 12.2 m in length.After harvest, the logs were air dried in sheds for six months to an approximate equilibrium moisture content of 19 %.After air drying, each log was manually debarked.Each debarked log was visually marked off and cut into several (straight) segments.Each log segment was shaved into a tapered, round segment using a 2.4-m, heavy-duty wood lathe.
A heavy-duty, fi nger-joint shaper was used to cut fi nger joints into the shaved log segments at one or both ends of each segment.The cutter of the shaper contained 15, two-wing, cutter blades.Each cutter blade had a tip angle of 10.7 degrees.The length, tip thickness, and pitch length of the blades measured 28, 1.588, and 7.938 mm, respectively.All fi nger joints at one end of each log segment were cut in one pass through the cutter.
Tapered, fi nger-jointed logs were fabricated using shaved segments of decreasing circumferences (from bottom to top).Segments were consolidated into a tapered, fi nger-jointed log in a 12.2-m long fi ngerjoint press using a resorcinol phenol formaldehyde resin at 506 g/m 2 .Each fi nger-jointed, tapered, roundwood log had a butt-end diameter of 15.5 cm and a length of either 9.1 or 9.6 m at test.The taper of all fi nger-jointed logs was 2.3 degrees, which equals the standard pole log taper required by the American National Standard Institute (ANSI) (ANSI, 2008).
Thirty six tapered, fi nger-jointed segmented logs were made.Nine laminated utility poles, each consisting of four of the fi nger-jointed logs, were fabricated and tested.Six of the nine were reinforced with bamboo strips, while three were not.Locations of the fi nger joints in the segmented logs were different for the four logs comprising a laminated utility pole, so that at most one fi nger joint would appear in any cross section of a laminated utility pole (Figure 1).Data categorized by fi nger joint location is not presented in this paper because an insuffi cient number of test samples could be obtained from upper locations in the pole due to taper.
Of the four fi nger-jointed logs comprising a laminated utility pole, three of the logs consisted of fi ve fi nger-jointed segments, while one log consisted of four fi nger-jointed segments.The minimum distance between fi nger joints on two logs comprising a side surface of a laminated utility pole was 0.6 m.
To construct all laminated utility poles, two adjacent fl at surfaces were cut into each fi nger-jointed, tapered log using a portable bandmill.After the fi rst fl at surface was cut, each log was turned 90 degrees, and the second fl at surface was cut.One additional fl at surface was cut into each of the four fi nger-jointed logs that were to form a bamboo reinforced laminated utility pole.Before gluing, the fl at surfaces of each log were sanded using a hand-held sander equipped with 100-grit sandpaper.This removed the saw-blade marks and smoothed the surfaces.Using a roller spreader, the same type resorcinol formaldehyde resin that was used to glue the fi nger joints together was applied evenly to the two fl at surfaces of each of the four logs that were to be pressed into an unreinforced laminated utility pole.These four glued pole logs were assembled and consolidated into a single unreinforced pole in a cold press (Figure 2).
Of the seven composite poles elected for fi nger joint test in this study, Poles 1 to 3 were made of wood only (without bamboo strip reinforcement), while Poles 4 to 7 were reinforced with bamboo strips.As shown in Table 1, Poles 1 to 3 produced more bending samples (fi nger-jointed and solid wood) than Poles 4 to 7. Since some materials were removed from the logs cut from bamboo strip-reinforced Poles 4 to 7, they usually produced less fi nger joint and solid wood samples than the composite poles without bamboo strip reinforcement.In addition to bamboo reinforcement, the number of bending samples obtained from each pole is also dependent on the diameter of the fi nger joint logs in the poles, and the failure locations and taper of the poles in the bending tests.Therefore, there is substantial variation in the number of bending samples obtained from the seven composite poles.
A fi nger joint in a log usually consisted of two log sections obtained from different trees.As a result, the strengths of the two pieces of wood consisting of a fi nger joint sample are usually different.To correct for the difference, solid wood samples were removed from the segments on both sides of a fi nger-jointed segment and were used as a control in the strength analysis of the fi nger joint.
After testing the poles in a cantilever fashion in accordance with ASTM D 1036-1999 (ASTM, 1999), Poles 1 to 7 were used to measure the fi nger joint strength of the poles.Each selected laminated pole was split along the glue lines of the pole into four fi ngerjointed pieces (Figure 3).Solid wood and fi nger joint bending samples from these logs were obtained as follows.A 1.3m segment was marked off at each fi ngerjoint location along the entire length of each fi ngerjointed log with the fi nger joint being located at the midpoint of the segment.The segment location in a fi nger-jointed log (1 to 5 from the bottom to the top) was labeled.After removal, each 1.3m segment was cut into three 42 cm sections, which were labeled as Segments A, B, and C, respectively, in Figure 4. Segment B, which contained a fi nger joint in the middle, was used to measure the strength and stiffness of the fi nger joint, while Segments A and C were used to measure the strength and stiffness of solid wood (without joints).The failure modes of all samples were recorded and will be reported in a separate manuscript.
Bending samples 46 cm long by 2.5 cm wide by 2.5 cm thick were then cut from Segments A, B, and C. Each segment was fi rst cut into 2.5 cm thick boards.Each board was trimmed to remove any curved surfaces (obtain true shaped boards for the bending samples).Two to fi ve contiguous bending samples were removed from the board, depending on the width of the board.Samples having defects in the wood, such as splits and knots in the middle of the samples, were discarded.The total number of usable bending samples obtained from each composite pole is summarized in Table 1.All test variables and their levels are shown in  Prior to bending testing, samples were conditioned in an air-conditioned room maintained at 21 °C for 5 weeks.The annual growth rings were counted on both ends of each fi nger joint sample.Of the two ring counts obtained from both ends of a fi nger joint or a solid wood sample, the mean value was used as the number of annual growth rings of the sample.Of all the fi nger joint samples obtained in a pole, half of the samples were loaded with the cross head parallel to the joint fi ngers (hereafter referred to as parallel orientation) and the other half were loaded with the cross head perpendicular to the joint fi ngers (hereafter referred to as perpendicular orientation).All samples were tested in static bending and loaded to failure on an Instron testing machine according to ASTM D143-94 (ASTM, 1994).All samples were loaded continuously throughout the test at a movable crosshead rate of 1.3 mm/min on a 360 mm loading span.
After testing, a 2.5 cm section was immediately cut from each sample near the point of failure and was used for moisture content (MC) measurement.The section was weighed and then put in an oven at 103±2 °C for 24 h.Each section was weighed again after drying, and the MC of each sample tested was calculated.The density at the time of testing was determined.
In the data analysis, the strength and stiffness of the small clear samples cut from segments A and C (Figure 4) were pooled and their strength average was used as the solid wood control to the strength of fi nger joint samples obtained from Segment B. Analysis of variance (ANOVA) was adopted to analyze the bending strength and stiffness data using Model 1 given below, Model 1: ) where y ijk denotes MOR/MOE, μ is the overall mean, α i is an effect due to poles, β j is an effect due to sections where the samples were cut, γ k is an effect due to loading directions, αβγ ijk is an effect due to the interaction between blade profi le and joint orientation, and ε ijk is the random residual error.Analyses were carried out using the GLM procedure of the SAS software computing system (SAS 2010).A signifi cance level of 0.05 was used for each statistical analysis.

REZULTATI I RASPRAVA
Numeric values for density, growth ring counts, MC at test, and moduli of rupture (MOR) of the fi nger joint and solid wood samples tested in this study are given in Table 3, while the moduli of elasticity (MOE) of the samples are given in Table 4. Also enclosed in Tables 3 and 4 are the MOR and MOE averages, respectively, for the composite poles from which the fi nger joint and solid wood samples were cut.The MOR and MOE values of laminated composites poles in Tables 3 and 4 were obtained from a previous study (Piao et al., 2013).Except for Composite Poles 2 and 5, the   bending strength averages of composite poles were lower than the bending strength averages of small clear solid wood samples cut from these poles.However, except for Composite Pole 1, the bending strength averages of composite poles were higher than the bending strength averages of fi nger joints in the poles, regardless of fi nger joint orientations (parallel or perpendicular) in the test.These results indicate favorable strength of a composite pole that is made of fi nger-jointed small diameter timber.
As expected, of all the seven composite poles tested in this study, the solid wood samples gave signifi cantly higher (p =< 0.0001) bending strength than fi nger-jointed samples, regardless of fi nger joint orientations in the test (Table 3).The bending strength averages of fi nger joints tested in the parallel orientation, fi nger joints tested in the perpendicular orientation, and solid wood samples were 36.1, 32.8, and 42.3 MPa, respectively.Finger-jointed samples tested in the perpendicular orientation (32.6 MPa) yielded the lowest bending strength but were not signifi cantly lower (p = 0.0939) than samples tested in the parallel orientation (36.1 MPa).The strength of the fi nger joints ranged from 26.9 MPa to 42.7 MPa when tested in the parallel orientation and 26.1 MPa to 40.0 MPa when tested in the perpendicular orientation (Table 3).Standard deviations were used instead of range of values.The ANO-VA showed that both the pole effect and test orientation signifi cantly affected (p =< 0.0001) MOR.The effect of section locations of fi nger joints along the poles was not signifi cant (p = 0.1726).There was no signifi cant difference for MOR among the fi nger joint locations.
It is important to know the strength of fi nger joints as compared to the strength of solid wood.Table 5 shows the strength percentages of fi nger joints to solid wood in both orientations.Of the seven poles tested in this study, the strength of fi nger joints in the parallel orientation was 76.6 to 97.8 percent of the strength of solid wood, while the percentages were 66.1 to 95.5 for fi nger joints in the perpendicular orientation.The strength averages for parallel and perpendicular orientations were 85.6 and 78.4 percent of the strength of solid wood samples, respectively.The overall strength percentage of fi nger joints (both orientations) to solid wood was 82.0.
The MOE results of the poles are presented in Table 4.The ANOVA found that pole and test direction main effects were marginally signifi cant (p = 0.0390 and 0.0206, respectively).The MOE for solid wood was lower than parallel jointed MOE (p = 0.0057).Table 4 also shows that the MOE of composite poles was substantially higher than the MOE of small clear samples, regardless of fi nger jointing or not.This is partly attributable to the resorcinol phenol formaldehyde (RPF) resin used in this study.In the jointing of both composite poles and fi nger joints, RPF was used as resin to bond wood members together.The cured RPF resin usually has a higher stiffness that the wood used in the study (Piao et al., 2005).The MOE did not signifi cantly vary based on pole locations in the fi nger- jointed logs (p = 0.2054).In addition, these differences can be partly attributable to the addition of the bamboo strips on the composite poles, which enhanced strength properties over poles without bamboo strips.Finally, due to the well known pattern of increased density from pith to bark in southern pine trees, the poles contained the highest density on the surface, which is critical because the surface properties are essential for determining the overall strength and stiffness of any member under static loading.

ZAKLJUČAK
Finger-jointed and solid wood small clear samples cut from seven composite utility poles were evaluated for their strength and stiffness properties.The strength and stiffness of the fi nger-jointed small clear samples were compared with those of small clear solid wood samples and those of composite poles made of small diameter timber.The bending strength of composite poles was less than the strength of the solid wood samples but greater than the strength of fi ngerjointed samples cut from the poles.However, the bending stiffness of the composite poles was substantially higher than the bending stiffness of both solid wood and fi nger-jointed samples cut from the poles primarily due to a number of factors including the RPF resin that was used to bond fi nger joints and fi nger-jointed logs in the poles, inclusion of bamboo strip reinforcement of the poles, and a likely higher percentage of mature wood in the poles as in the small test samples.Fingerjointed samples tested in a perpendicular orientation yielded the lowest bending strength but were not signifi cantly lower than samples tested in a parallel orientation, indicating that joint orientation was not an important issue regarding the strength of the poles.The strength averages for parallel and perpendicular orientations were 85.6 and 78.4 percent of the strength of solid wood samples, respectively.The overall strength percentage of fi nger joints (both orientations) to solid wood was 82.0.This study has demonstrated that small diameter timber can potentially be used to make fi ngerjoint logs for the production of laminated utility poles.

Table 1
Number of solid wood (A and C) and fi nger joint (B) bending samples obtained from laminated composite poles made of small diameter timber