Lateral bracing to prevent buckling of compression members is a common design requirement in wood frame engineering. The obvious example of a compression member is a post, but there are other more numerous components where compression occurs and necessitates analysis of bracing. In particular we will look at truss chords and webs, and pre-engineered wood I-joist flanges.
In a truss, compression members can occur throughout, including both chords and webs. The strategy employed by truss design software to analyze lateral brace spacing is generally that used in post design. Resistance factors are used to account for brace spacing. In Canada, the process to follow in calculating compressive resistance is outlined in the Wood Design Manual's section on Compression Members. In it, the spacing of lateral braces influences the total resistance by affecting both the KZ factor and the CC factor (which is a subcomponent of the KC factor).
In the Mitek Engineering software we can see the calculation at play. For chords, an exact lateral brace spacing is provided as an alternative to sheathing. For a web under significant compression, the software will specify that the span must be divided evenly by the presence one or more lateral braces (unless t-braces or scab braces are enabled instead). Figure 1 shows an example of Engineering's brace spacing calculation for a truss's 2x4 SPF #2 top chord.
An I-joist flange is in compression when the joist is experiencing a bending moment in the direction of that flange--for example, the top flange of a uniformly-loaded simple-span joist, or the bottom flange of a uniformly-loaded multi-span joist near a mid bearing. While a top chord is usually laterally-supported by sheathing, a bottom chord is frequently bare and therefore lateral support becomes a design concern. The joist's greatest negative moment (and the bottom chord's greatest compression) occurs right at the bearing where the chord is laterally supported by toenailing; however the moment in the regions adjacent to the bearing can also experience high negative moment/compression. Design software typically offers a mechanism for analyzing the required lateral support for such compression, but the results of the calculation are typically relegated to calc sheets as opposed to being properly displayed on joist layouts. As a result this bracing is one of the most overlooked elements in joist design.
The approach carried out in the Canadian versions of design software is typically that found in the American National Design Specification For Wood Construction, which since the 2012 version has recommended designing an I-joist compression flange as if it were a column, and using the CP factor of the column design for the CL factor of the joist's moment capacity calculation. Software programs generally carry out at least the latter step--but with the approximately equivalent factors from Canadian design procedures. The KC factor from column design is used in place of the KL factor of moment design (but still referred to by the American CL).
In the now-defunct CS-Build/CS-Beam software (originally called Keymark), the user can enter a brace spacing or tell the software to specify the maximum spacing. We will inspect a CS-Beam report for a joist with the maximum bracing option enabled. The theory employed by the algorithm is that bracing is only required where the joist is actually under negative moment: a region on either side of each mid-bearing, extending outward to the point of zero moment where the chord can be thought of as braced. The moment capacity of the joist with the brace spacing set to the length of this region is first tested against the maximum negative moment, and if it will suffice, no additional bracing is needed. If it will not suffice, the capacity is calculated with increasingly smaller brace spacings until one is found to produce adequate capacity. This process is repeated for all load cases. Figure 2a shows the section of the report for a load case where the region length would be a sufficient brace spacing. Figure 2b shows the section from the same report showing a load case where smaller brace spaces need to be tested.
The iStruct software deals with bottom chord bracing in a similar fashion. A checkbox in the program settings defines whether the points of zero moment can be used to determine the unbraced length of the chord before the application of bracing. If this setting is turned off the software will always use an entire span from one bearing to the next. iStruct uses a Ke value of 1.0. Figure 3 shows the calculation of CL for two load cases of the same joist. iStruct does not report the E05 and fC values of flanges, but for this joist it seems to be using a FC/E' value of 7.6 x 10-5.
Unfortunately the Mitek Sapphire software, despite being a modern design program in most ways, does not show the user its logic for bottom chord bracing. As of version 8.5.3.233.Update6.12 the longest span of the joist is always reported as the brace spacing, at least in reports generated using the default template.