1. Introduction
The research presented here was conducted on Southern Pine by Ralph L. Rupert and Marshall S. White from Virginia Tech’s Center for Unit Load Design. I want to recognize and thank the people at Potomac Supply Corporation for sponsoring and funding this valuable research, as well as our friends at Virginia Tech who worked so hard gathering and analyzing data on heavy duty Southern Pine pallet stock. The information contained in this article came from a more complete, technical report published by Virginia Tech’s Center for Unit Load Design. While the research was done on heavy duty Southern Pine block pallet cut stock, the overall result is likely to apply equally well to other species of softwood pallet parts for both block and stringer pallets. Extending the research results logically, similar results should hold for hardwood pallet parts as well.
Most pallet specifications do not designate the quality of raw material used in fabricating the pallet’s components. Specifications typically focus on the pallet parts themselves, not the raw material from which they are remanufactured. It makes sense, however, that raw material selection influences the distribution, by size and frequency, of growth related defects within components that are selected based on maximum allowable defect size. Therefore, the mechanical properties of these pallet parts will depend on the raw material quality. The purpose of this study was to determine the effect of the American Lumber Standards Committee (ALSC) National Dimension Lumber Grade (NGR) on the flexural strength and stiffness of Southern Pine pallet parts used in a heavy duty Southern Pine block pallet design. The issue behind this research concerns whether or not pallet parts that are manufactured from lower grades of lumber and then sorted according to a quality criteria differ in performance from parts that are manufactured from higher grades of lumber and then sorted according to the quality criteria.
This study confirms that pallet parts manufactured from higher grades of Southern Pine are superior in bending strength and stiffness to pallet parts that are sorted according to the same grades after being manufactured from lower grades of softwood.
2. Study’s Objective
• Evaluate the effect of Southern Pine Dimension NGR Lumber Grade on the bending strength (Modulus of Rupture, MOR) and stiffness (Modulus of Elasticity, MOE) of a heavy duty 48×40 block pallet’s components.
• Evaluate the effect of mill source on flexural strength and stiffness of block pallet components.
• Evaluate the effect of finger jointing on SYP block pallet component flexural strength and stiffness.
3. Test Materials
NGR dimension Southern Yellow Pine grades 1, 2, and 3 were supplied by ten different mills for testing. The lumber was regraded by Timber Product Inspection (TP) of Conyers, GA at Potomac Supply Corporation in Kinsale, Vir. to verify the NGR grade assigned by the mill source. Block pallet connector boards, interior top deckboards, bottom base boards and bottom butted boards (uncut – 40” length) were manufactured for testing. These parts were graded by TP staff to assure conformance with a specific block pallet part grade criteria. Ten (10) replicates of each pallet part from each mill and NGR grade were tested, for a total of 1200 test specimens.
An additional set of finger jointed pallet parts was furnished for testing. Twenty-five connector boards, interior top deckboards, and bottom base boards were tested. The joints were produced from commercial, long length, fingerjointed dimensional lumber and cut into the specified block pallet parts, which produced randomly placed joints in the various pallet parts with many having multiple joints. The rounded tip finger joints were produced perpendicular to face with a 4.2 length to pitch ratio.
4. Test Methods
All pallet parts were conditioned to a target 12% equilibrium moisture content. The pallet parts were tested in third point bending using an MTS servo-hydraulic testing machine. The test span for 40-inch deckboards was 36 inches, while the test span for the 48-inch connector board was 44 inches. Deckboards were tested at a deformation rate of 3.2 inches/minute, with load and deflection measured continuously during each test.
The moisture content was obtained by cutting representative samples from each test specimen at the time of testing and using the oven-dry method. The moisture content in percent is the (green weight of specimen – dry weight of specimen / dry weight of specimen) x 100. The specific gravity, calculated as (oven-dry weight/oven-dry volume), was determined using oven-dry wax covered specimens immersed in water to measure volume.
The flexural modulus of rupture and modulus of elasticity were calculated after adjustments for shear effects and rate of loading, then corrected to 12% moisture basis. Actual moisture contents ranged from 8 to 14%.
5. Test Results
The following sections discuss the test treatments of NGR lumber grade, source of lumber supply, and effect of finger jointing.
5.1. The Effect of NGR Lumber Grade on the Properties of Southern Pine Block Pallet Parts
Pallet parts, sorted by part and grade, were tested according to the modulus of elasticity (MOE), modulus of rupture (MOR), and density of the pallet parts tested. Statistical analysis determined that observed differences in numbers were not large enough to be statistically different; in other words, any differences are likely to be due to natural variation. There is no detectable statistical difference (at a 95% confidence level) between the average MOR and MOE of the different pallet parts within each NGR lumber grade. In other words, when testing cut stock that was manufactured from one of the specific lumber grades, any variation in the data is within the range of normal expectations. This is not surprising for lumber that has been uniformly graded. Thus, the remaining discussion will focus on NGR grade affects only (the differences between one grade and another).
Table 1 contains the test results of pallet parts sorted according to the lumber grade from which they were produced. Table 2 shows the 5th percentile values (a statistical, calculated value where 95% of the values are greater) of pallet parts sorted by NGR grade. Figures 1-4 are bar charts of the average and 5th percentile MOR and MOE and show the percentage difference compared to NGR Grade 3 at 100%. Statistical analysis (ANOVA) at a 95% confidence level indicates there is a significant treatment effect. In other words, the average bending strength and stiffness are statistically significantly different from one lumber grade to another. The average flexural stiffness of parts (Figure 1) produced from NGR Grade 3 lumber is 21% less than from NGR Grade 1 and 7% less than from NGR Grade 2 lumber. The average flexural strength of parts (Figure 2) produced from NGR Grade 3 lumber is 30% less than from NGR Grade 1 and 11% less than from NGR Grade 2 lumber.
The variation of pallet part flexural strength and stiffness increases as raw material grade decreases, thus leading to larger differences in the lower 5th percentile. The lower 5th percentile flexural stiffness of pallet parts (Figure 3) from NGR Grade 3 is 50% less than from NGR Grade 1 and 32% less than from NGR Grade 2. The lower 5th percentile flexural strength of pallet parts (Figure 4) from NGR Grade 3 is 47% less than from NGR Grade 1 and 21% less than from NGR Grade 2. This, in spite of the fact that the pallet parts were graded to the same minimum quality criteria.
The average density of pallet parts from NGR Grade 1 lumber is statistically higher than from NGR Grades 2 and 3; however, the average density of pallet parts from NGR Grade 2 is not statistically higher than pallet parts from NGR Grade 3.
5.2. The Effect of Source of Supply of Dimension Lumber on The Properties of Southern Pine Block Pallet Parts
The average flexural stiffness and strength of pallet components produced from dimension lumber were statistically different from one mill source to another according to an analysis of variance test (ANOVA). The Analysis of Means (ANOM) test compared the average flexural strength and stiffness values of each mill to the grand average of all the pallet components to determine which of the mill sources were different. Any lumber buyer who purchases lumber from a number of different sources can verify that there are quality differences from mill to mill, even when buying supposedly comparable grades. The same is true for pallet cut stock that is produced from the three different grades of southern pine. There is a 20% difference in both the average flexural strength and stiffness of pallet parts from the highest and lowest of NGR graded lumber.
Strength and stiffness are highly correlated with wood density. The test specimens with the highest average MOE also are the densest, while the specimens with the lowest average MOE have the lowest average density. The differences in MOE, MOR, and density between the sources of NGR graded lumber may be attributed to any one or a combination of differences in quality within grade, differences of sawing patterns, or Southern Pine species mix. The exact cause of the source influence was not determined and was not within the scope of the study.
5.3. Relative Strength and Stiffness of Finger Jointed Heavy Duty Block Pallet Parts
The average flexural strength and stiffness by NGR grade, including the finger jointed samples, are shown in Table 1 with the lower 5th percentile values shown in Table 2. When compared to pallet parts from NGR Grade 3, the average flexural stiffness (MOE) of the finger jointed pallet parts was 7% lower and the average flexural strength (MOR) of the finger jointed pallet parts was 11% lower, while the lower 5th percentile flexural stiffness (MOE) of the finger jointed pallet parts was 5% greater and the lower 5th percentile flexural strength (MOR) of the finger jointed pallet parts was 25% greater. The relatively better lower 5th percentile values is attributable to the lower variability of the finger joint components.
It should be noted that all of the test samples broke at a finger joint that was near or between the load applicators. While several boards contained multiple finger joints, failure did not occur at those finger joints located outside the load applicators. Glue line failures were not observed. Some samples did propagate the split into surrounding solid wood, but most failures were within the boundaries of the joint indicating that gluing quality was acceptable. The finger joints clearly act as an anomaly which influences test results. This leads to less variation in strength compared to non-finger jointed parts, thus the lower 5th percentile strength of the finger jointed parts is similar to parts from NGR Grade #2 lumber and better than parts from NGR Grade #3 lumber.
6. Summary and Conclusions
• The ALSC National Grade Rule Dimension Lumber grade significantly influences the flexural strength and stiffness of heavy duty block pallet parts graded to the same criteria and sawn from the lumber.
• The source of supply of the Southern Pine NGR graded dimension lumber significantly influences the strength and stiffness of the heavy duty block pallet parts by as much as 20%.
• The average flexural strength and stiffness of the finger joint pallet parts tested is lower than the average flexural strength and stiffness of similar quality pallet parts sawn from NGR Grade #3 components.
