Field Data Required for Use of the Grading Protocol
Unfortunately, the wealth of information described above often eludes the practitioner who must make a decision about the structural integrity of a building. It is the goal of this project to provide the means for an engineer to use the current grading rules to make an informed decision about allowable design values for wood members in historic buildings. A simple query-based wood grading database, referred to as the wood grading program, was developed in order to provide engineers with the tools necessary to conduct a basic wood grading assessment for timbers within historic structures. To simplify the field data required as input for the program, only the wood species, member thickness, and width of the structural elements in question need to be determined.
Knots and slope of grain are the most common and most limiting of lumber defects (Cramer et al., 1996). Therefore, the grading program uses just these two factors to determine appropriate grade designation for one of five wood species. Prior to utilization of the wood grading program, therefore, a complete wood condition assessment should be conducted to determine the presence and/or significance of any other grade-limiting defects.
To determine the appropriate species, a sample of each type of structural member under consideration for the grading protocol must be removed and sent for analysis. Generally, the same wood species is used for identical framing members such as joists or rafters, however, species may vary between framing member groups (e.g. the rafters may be southern yellow pine, but the joists may be eastern white pine). Samples can be sent to a number of private consultants for a fee or to a public or government institution such as the U.S. Forest Products Laboratory Center for Wood Anatomy Research that can provide species analysis free of charge.
Samples should be taken from sound wood and should measure a minimum of ¼-inch wide x ¼-inch deep x ½-inch long. The soundness of the wood sample can be determined by rolling it between the fingers; if the wood breaks apart, it should not be submitted and a new sample will need to be taken. To extract a sample, use a sharp knife, craft saw blade, and/or a chisel to make two cuts across the grain of the element. These two cuts should be a minimum of 3/16-inch deep and ½-inch apart. A specimen can be split out by prying up at one of the incised points with a knife, or if a chisel is used, the edge of the chisel can be placed in one of the cuts and then angled down the grain towards the other cut. A sharp tap with a small hammer should provide enough force to remove a good specimen from the wood member. All samples should be taken from an inconspicuous spot on each wood member, and placed within a labeled bag for accuracy. It should be noted that a variety of species can be used within a single building, so at least one sample should be taken for every structural member type (rafters, joists, columns, etc.) under consideration for application of the wood grading protocol.
Knots and Their Measurement
Knots are generally considered the most significant of the numerous strength- limiting defects occurring in lumber. Three major strength-reducing effects arise from the presence of a knot: part of the board cross-section is reduced as harder, denser, but structurally weaker knotwood takes the place of the regular wood fibers; a stress concentration and subsequent reduction in capacity is induced by the material inhomogeneity of the knot surrounded by the rest of the board, and lastly, the growth pattern of the trunk is disrupted by the branch causing the knot, which results in considerable distortion of the grain angle around the knot (Cramer et al., 1996). This grain angle distortion can allow for the development of tensile stresses perpendicular to the grain and the formation of checks and microfractures as the wood dries (Cramer et al., 1996).
The location of the knot has an impact on member strength. Therefore, within the grading program are three knot size limitations for knots based on the location of the knot. Centerline knots on the wide face of an element have the least impact on grade and therefore have the largest allowable knot size. Edge knots on the wide face generally increase localized tensile stresses, and therefore have smaller knot size limitations. The same holds true for knots on the narrow face of an element. For some smaller elements (2 to 4 inches in thickness), knots on the narrow face are considered to be identical to edge knots on the wide face. While the location of a knot along the length of a bending member (within the middle third of the length or in the outer two-thirds) also affects the performance of a beam, this was not taken into account to provide a level of simplicity to the field measurements. This results in a conservative limitation on knot size for the outer third of beams.
The measurement of knots varies depending upon the species, size and function of the element containing the knot. For example, knot measurement on columns under axial loads is different than knot measurement on beams and joists in bending. The following excerpts and figures provide a guideline for the different methods of knot measurement.
From Western Lumber Grading Rules 05 (Western Woods Product Association, 2005, used courtesy of Western Wood Products Assn.):
In all Framing lumber 4” and less in thickness, the size of a knot on a wide face is determined by its average dimension as in a line across the width of the piece. The size of knots on wide faces maybe increased proportionately from the size permitted at the edge to the size permitted at the centerline. Knots appearing on narrow faces are limited to the same displacement as knots specified at edges of wide faces. Knots in Beans and Stringers and Post and Timbers are measured differently than knots in 4” and thinner material. Examples of these measurement methods are shown in Sections 212.00 and 213.00
Visible knots on lumber or timbers can be measured using an acetate sheet with a ½-inch grid to facilitate measurement (Figure 5). As shown in Figure 2, for structural joists and planks, the size of a knot on the wide face is determined by measuring the knot size on both wide faces and calculating the average knot size through the piece. Knots on the narrow face are assessed in a similar fashion as edge knots on the wide face and are measured with the acetate grid parallel to the long axis of the piece. Knots on larger members are measured differently, as shown in Figures 3 and 4.
For structural joists and planks, and beams and stringers, the grid should be placed parallel to the edge of the timber and the knot dimension measured parallel to the long axis of the timber. For posts and timbers, the grid should be aligned across the narrowest face of the knot and should be measured between parallel lines edge to edge, from the most distinct point of grain deviation. In other words, the acetate grid may not be parallel to the edge of posts and timbers as it is for beams and stringers.
Occasionally, the knot, which is the remnant of a tree branch, has a very distinct boundary, but this is typically not the case, and the measurement is subjective. Because of the firmness of knots (due to denser wood than that surrounding the knot), differential drying of dense knotwood often causes knots to become raised from the normal surface of a timber; hence, knots on painted timbers often “telegraph” through the painted surfaces and are visible for measurement. Also, radial checks may be useful in defining the knot boundary as they will not extend into the surrounding wood.