Capacity of oriented strandboard joints with overdriven nails.

Abstract

This article presents the results of a testing program to evaluate the capacity of oriented strandboard joints with overdriven nails. Several single nail wood joints were tested using a cyclic loading procedure. Joints were assembled with 11 mm sheathing panels, 38 by 89 mm wood members, and gun-driven 2.9 mm diameter by 60.3 mm long nails. Nails were flush-driven and overdriven 1.6, 3.2, and 4.8 mm. Results indicate that any level of nail overdriven depth will reduce the capacity of a joint. Compared to flush-driven joints, joints with nails overdriven 1.6, 3.2, and 4.8 mm experienced 9, 13, and 22 percent loss in load capacity, respectively. Approximate reductions in ultimate displacements were 22, 41, and 45 percent and 20, 40, and 44 percent in cumulative energy dissipation for joints with nails overdriven 1.6, 3.2, and 4.8 mm, respectively. Initial stiffness decreased 20 percent for joints with 1.6 mm overdriven nails but increased 8 and 19 percent for joints with 3.2 and 4.8 mm overdriven nails, respectively.

**********

Modern advances in construction technology have led to new methods, materials, and tools for increasing efficiency and profitability. One tool of the modern construction era is the pneumatic nail gun. Although the nail gun has revolutionized the timber framing industry, it can cause construction problems when used to connect sheathing panels to framing members. Figure 1 shows schematically the three conditions (flush, underdriven, and overdriven) to which a nail can be driven into the sheathing. Nail guns available on the market have some ability to be mechanically adjusted and pressure calibrated; however, they are often not adjusted successfully. One reason for this is the variation in framing member and sheathing properties, which cannot be accounted for by any type of adjustment or calibration. Newer nail guns, manufactured with a shorter driving pin, still do not account for variations in properties of framing members. Another cause for nail overdrive is apathy during construction. If nails are underdriven they interfere with subsequent wall coverings. To avoid having to hammer underdriven nails, which would undermine the automation and timesaving provided by using the gun, guns are usually adjusted and calibrated to ensure that nails will be slightly overdriven. In many cases, however, nails are significantly overdriven.

A survey of light-frame timber construction along the Wasatch Front, Utah was conducted to assess the prevalence of overdriven nails in new construction (Jones and Fonseca 2000, Rabe and Fonseca 2000). Only shear walls were considered in the survey. Every structure randomly encountered and surveyed was sheathed with 11 nun oriented strandboard (OSB). The sheathing nail type varied slightly, but approximately 90 percent of the surveyed walls were nailed with 2.9 mm diameter by 60.3 mm long gun-driven (8d) nails. Due to the overwhelming presence of 11 mm OSB and 8d nails, only that combination was included in the analysis of the data. A total of 21 locations were included in the survey. At each location, four rows of nails were randomly selected along sheathing panel edges and 10 consecutive nails were measured for driven depth along each row. A total of 840 nails were measured.

Figure 2 summarizes the results of the survey. Underdriven nail depths are shown as negative values. The distribution is fit with a normal density curve as a way to describe the distribution and as a tool for analyzing the significance of the findings. The mean value for overdriven nail depth is 2.0 mm. The normal distribution fitted to the data has a standard deviation (SD) of 2.1 mm. Statistical analysis of the data indicates a significant deviation of the mean from being flush: t = 27.5, df = 839.0, p = 0.0. That significance is proven using a matched pairs t procedure, which calculates the chances of having the sample mean deviate from the population mean. In this case, the mean of the sample (2.0 mm) is compared to the population of nails as a whole with a theoretical mean of 0.0 mm, which corresponds to the flush-driven nail condition. The corresponding t value is 27.5, which alone is significant since t does not reach values much higher than 3.5 before p-values become infinitesimal (Moore 1995).

[FIGURE 1 OMITTED]

[FIGURE 2 OMITTED]

Underdriving or overdriving nails is common practice along the Wasatch Front. More than 80 percent of the nails surveyed were overdriven, and approximately one-third of them were overdriven at least 3.2 mm. These findings are consistent with other surveys. Investigation of many woodframe multi-family buildings in the San Francisco Bay area, for example, revealed that about 80 percent of the nails in 9.5 mm plywood shear walls were overdriven, with many nails overdriven 3.2 mm or more (Gray and Zacher 1988). A less formal survey following the Northridge Earthquake discovered that "numerous" plywood shear walls had 50 percent of the nails overdriven (Hall 1996).

Objective

The objective of this article is to report on the overall capacity of OSB joints with overdriven sheathing nails. This study is companion to a previous study (Fonseca et al. 2006) and both are an ongoing effort to establish a database for analytical modeling. The difference between the two studies is the type of sheathing material: the current study uses OSB while the previous one used plywood. The research presented adds a significant set of variables to the general fund of tests. The data included herein will allow researchers conducting analytical studies to obtain results from their analyses that should be more accurate.

Previous research

Overdriven nails have been recognized as a design issue for many years; yet, only limited research has been conducted to investigate their effects on the capacity of nailed wood joints. Zacher and Gray (1985, 1989) and Gray and Zacher (1988) tested fifteen 63 by 71 mm small-scale specimens. Specimens were assembled with framing along all of the edges and multiple nails. Little information about the material used and construction of the specimens are given except that some specimens where assembled with CD plywood (with exterior glue) and some with Structural I plywood. Fastener type, spacing, overdriven depth, boundary conditions, and the loading location are also not described. Testing was conducted using a cyclic, displacement-controlled loading sequence at 2.0 Hz. Results and behavior of the specimens are not discussed. The conclusion was that the largest component of the measured distortions was due to fastener flexure and movement of the fastener within the framing rather than sheathing strains.

Andreason and Tissell (1994) also investigated the effects of overdriven nails by testing 27 small-scale specimens. Testing was conducted using a monotonic loading procedure. Each specimen consisted of three pieces of plywood connected to a wood member. Tests compared specimens with flush-driven nails to specimens with nails overdriven to two different depths. Various thicknesses of plywood were used. Ultimate loads were reduced by amounts ranging from 2 to 17 percent. Loads at a slip of 0.06 mm, considered as the proportional limit for lateral resistance of fasteners, were higher for specimens with overdriven nails that those with flush-driven nails, which was attributed to the increased panel density underneath the head of the overdriven nails. From their testing, Andreason and Tissell concluded that overdriving will not significantly affect the strength of a shear until at least 20 to 25 percent of the nails are overdriven, which represents a 3 to 4 percent reduction in ultimate load compared to flush driven nails.

Bao (2002) tested 15 plywood and 15 OSB specimens using the ASTM D1761-88 (ASTM 1988) monotonic load protocol. Tests considered two nail depths, flush-driven and 1.6 mm, in 9.5 mm sheathing panels. The results indicated an increase in 5 percent offset yield strength but no change in ultimate strength for the overdriven joints.

Fonseca et al. (2006) determined the strength of plywood joints with overdriven sheathing by testing several joints constructed with 12 mm plywood sheathing attached to 38 by 89 mm wood members using 8d common or 8d cooler nails. Four nail-drive depths were investigated using cooler nails: flush, 1.6, 3.2, and 4.8 mm; common nails were all flush-driven. The results indicated that the stiffness of plywood joints using cooler nails increased from 10 to 21 percent with an increase in nail-drive depth. Strength, however, was reduced 11 percent for the 4.8 mm overdriven depth. Moreover, with an increase in nail-drive depth, the displacement capacity reduced from 38 to 77 percent, and energy dissipation reduced from 16 to 74 percent. Plywood joints using flush driven common nails have similar mechanical properties compared to plywood joints using flush driven cooler nails, except joints using common nails have 36 percent less displacement capacity.

Except for the tests by Bao (2002), all of the tests were conducted using plywood specimens. Nevertheless, even Bao only tested 15 OSB specimens with one overdriven depth. Research conducted to date on specimens with overdriven nails is incongruous and the results are mixed. Clearly, due to the many variables affecting the behavior of specimens with overdriven nails, investigations conducted so far are simply not sufficient. Therefore, data presented herein will significantly augment the available database.

Experimental approach

Specimen and materials

Joints were built according to the dimensions shown in Figure 3. The placement of the nail was chosen to represent the condition encountered when the edges of two sheathing panels are attached to a single wood member with a gap of 3.2 mm between them (APA 1980). A nail edge distance of 9.5 mm was maintained for all of the joints.

Print Get the Mag Weekly Updates