Mobile Edition
Print
Get the Mag
Weekly UpdatesApproximately 20 percent of the hardwood lumber produced must be edged and nearly all lumber must be trimmed (Kline et al. 1990). Over-edging combined with improper trimming methods can result in a substantial loss of clear, merchantable wood as well as a drop in grade. A sawmill should optimize the processing of each log in order to remain competitive in today's hardwood market. To maintain a profit margin the sawmill must achieve maximum hardwood lumber value based on several factors, such as balancing between lumber volume and lumber grade (Kline et al. 1990). Many variables need to be considered during the edging process, including fluctuating prices, inaccurate estimates of board surface measure, operator training and fatigue, and combinations of edging and trimming solutions (Abbott et al. 2000). The sawmill must also understand that consumers can be demanding when it comes to the products they wish to purchase, which places restraints and strict goals on the mills. Some purchasers may want lumber that exceeds National Hardwood Lumber Association (NHLA) requirements (NHLA 2006). The sawmill may need to produce lumber edged on a more strict rule in order to fulfill a particular order.
It is evident that substantial losses can occur while the edging process takes place (Flann and Lamb 1966, Williston 1979, Bousquet 1989, Regalado et al. 1992). Several researchers have evaluated accrued sawing and edging losses. Williston (1979) found loss of a log's original volume could be up to 45 percent when converting slab boards and edgings into chips. Excessive edging practices have been reported to result in value losses of 30 percent (Bousquet 1989). Three hardwood sawmills were investigated and average losses of 32 percent of the lumber value and 10 percent loss of lumber volume were observed compared to optimal edging and trimming (Kline et al. 1993). Edger operators tend to be biased toward the removal of wane beyond what is necessary according to lumber grading rules (Lee et al. 2003). Under-edging can cause the board to be sent back for further processing, while over-edging can result from failing to consider the numerous possible combinations of grade and surface measure that can be produced from each board. Reduction of surface measure, due primarily to over-edging, can result in a significant monetary value loss. These losses become apparent when final grade and surface measure are determined.
Human error could result in value loss of 62 to 78 percent of what is optimum in lumber yields (Schmoldt et al. 2001). Therefore, much interest has been raised in developing automated scanning techniques to optimize the value of each board. For example, efforts have been made to advance technology in automated hardwood lumber grading systems (Klinkachorn et al. 1992, Kline et al. 2001). Research has clearly proven the advantages of automating edging/trimming operations to maximize lumber value, rather than volume (Schmoldt et al. 2001). Lee et al. (2003) described a prototype system that scans rough, green lumber and automatically provides an optimal edging and trimming solution along with resulting lumber grade. But, the cost of this equipment and the lack of total reliability of the system in detecting defects in rough lumber have made the adoption process slow in hardwood sawmills. Research surveys have suggested that wood processing is shifting toward semiautomation in Pennsylvania but basic production is still unchanged (Smith et al. 2004).
Hardwood sawmills are the most important hardwood roundwood consumers in West Virginia (Alderman and Luppold 2005). Sawmill size and production rates vary across the state. Luppold et al. (2000) stated that a third of the eastern hardwood lumber production is provided by mills that produce less than 3 million board feet (MMBF) annually. Mills in the northeastern region are more numerous, southern region mills are larger on average, and northwestern mills are smaller in production size and number of mills (Alderman and Luppold 2005). Most hardwood sawmills produce a wide variety of products for different end-use markets (Luppold 1995). These markets allow small-scale sawmills to remain competitive by producing customer-specific products. With high-quality logs becoming more difficult to find and a need to create new markets for low-grade lumber (Wiedenbeck et al. 2004), there is a definite need to research opportunities for optimizing profits of small-scale sawmills in the region. Optimizing profits from each board would greatly benefit sawmills and help ensure that these small-scale sawmills remain competitive in the hardwood lumber market. Therefore, the objectives of this study were to: 1) assess the efficiency of small sawmills in West Virginia by solely using NHLA grading rules; 2) compare the mill grade of a board with the assigned NHLA grade from a private grader both prior to and after edging and trimming; and 3) examine if the effort of assessing each face of a board prior to edging would be beneficial in the edging decision-making process.
Materials and methods
Board measurements
A total of 360 boards of five species including red oak (Quercus rubra), white oak (Quercus alba), red maple (Acer rubrum), black cherry (Prunus serotina), and yellow-poplar (Liriodendron tulipifera), were assessed in six sawmills across West Virginia between June and September 2006. The objective was to assess two species at each mill on separate occasions: however, measurements of two different species at two of the mills were unable to be done. Therefore, samples from four of the six sawmills consisted of 30 boards each of two different species. The remaining two mills sawed the same species on both occasions and those species were used for the study. Unedged boards were gathered directly after being sawn from a log, which enabled measurements for the piece before any further processing. Unedged boards were collected randomly, but generally contained wane on two edges. Only unedged boards that were going to be sent to the edger were examined.
A portable stand was developed in order to imitate the edging process and properly measure each unedged board. The stand was equipped with sliding laser lines to determine widths. A National Hardwood Lumber Association (NHLA) certified lumber grader was employed to determine the grade of each face of the boards for comparison to the mills certified lumber grader. Once the NHLA grader determined the maximum width of the board, a grade and surface measure were measured and recorded. The locations of the laser lines were marked on the end of the flitch to assure proper alignment on the second face for grading purposes. The hired NHLA grader was asked to process the board in order for it to produce the highest price. This requires a combination of surface measure and grade. Prices were acquired from each mill in order to determine optimum edging lines.
Data were collected for each unedged board to determine the optimal grade and resulting surface measure. Unedged boards were placed on sawhorses for stability, while a reference line was attached at each end of the unedged board on one edge with the top face facing up. The top face is the outer face when sawn from a log, while the bottom face refers to the inner face of the board sawn. The reference line served as the point for all width and defect measurements. Width measurements were taken at l-foot increments along the unedged board starting at zero feet and ending at the nearest half inch on the opposite end. Width measurements consisted of the distances from the reference line to the edge of the wane or clearwood, end of clearwood, and end of wane. Four separate measurements were generally taken, two for wane on either edge and two for clearwood, while three measurements were necessary if only one edge contained wane. These measurements allowed for a good representation of the amount of clearwood as well as the amount of wane on each edge of the board, which also allowed for the analysis of how much wood was removed during the edging process.
Defects were examined for each board prior to grade and surface measure measurements. The defect data included type, length, width, and location on the board. Defects were classified as sound and unsound knots, splits, holes, stain, worm holes, dog marks, shake, miscut, pith, and decay. If the end of a board was uneven, then a defect on the board was defined as miscut, which was caused during felling of a tree and requires trimming to square the board end. Defects were measured in a similar manner as wane and board width measurements. For example, lengths were recorded for the beginning and ending points of a defect. A minimum of two widths were measured from the reference line to determine maximum width of a defect. In some cases, such as splits, more than two widths were recorded to accurately measure the width of the split for its entire length.
In order to make the measurements consistent, the same NHLA lumber grader assessed and inspected each flitch to determine maximum surface measure and the highest grade. By adjusting simulated saw lines, the maximum width was determined for the top face and then each face was graded upon the determined width and length of that flitch. The NHLA lumber grader assessed each face of the flitch, which allowed him to determine a grade for each face. The surface measure was recorded, as were the grades for both faces. Each flitch was then edged and trimmed, and a grade given by a sawmill lumber grader was recorded. Each board was then reassessed by the NHLA lumber grader, including the determination of the final surface measure, top face and bottom face grade, and a final grade for the board.
Analysis
Analysis was performed to determine if the sawmills were optimizing each board by comparing grades of top and bottom faces and surface measures of pre and post processing. Specifically, the edging and trimming practices were assessed based on initial measurements performed by the NHLA grader and the same measurements taken after the board was edged, while the grading was assessed by comparing the grade by the mill grader with the grade given by the NHLA grader.
FEED