- The objective of this research is to use a three-dimensional (3D) analysis
method to include 'system effects' directly in the design of light-frame roof truss
assemblies. The goal of this study is to develop an improved and practical method to
determine the structural performance of 3D roof truss assemblies used in residential
In this study, three types of 3D truss assemblies (i.e., T-shaped, L-shaped, and
a complex assembly) are investigated. T-shaped, L-shaped, and complex assemblies
are composed of 4 (total of 37 trusses), 17 (total of 56 trusses), and 27 (total of 123
trusses) different types of individual trusses, respectively. The VIEW (Visually
Integrated Engineering Window) program, used by a truss plate manufacturer to
design truss assemblies, is used to layout assemblies and to design individual trusses.
The program uses a conventional design procedure (CDP) by analyzing one truss at a
time in two-dimensions. A commercially available structural analysis program,
SAP2000, is then utilized to model 3D truss assemblies with a system design
procedure (SDP). The structural responses including CSI, truss deflections, and
reactions from both CDP and SDP are compared and the 'system effects' are
From this investigation, there are three system effects observed by the SDP,
but not accounted for by CDP. These effects were observed in all three assemblies.
The first is the reduced applied load effect. This effect occurs because the CDP
assumes a 2-ft spacing for every truss in an assembly, whereas the SDP assumes the actual tributary area, with normally a lower total applied load. In most cases, the CDP
overestimates the applied loads. For example, in the complex assembly model, the
total applied load assumed by the CDP was 155000 pounds (assuming 2-ft spacing on
center for every truss in the assembly) while the total load for the SDP was only
142000 pounds (based on the actual overall roof loading area in the assembly). The
second system effect is deflection compatibility. Unlike the CDP, SDP provides
deflection compatibility because it analyzes the entire 3D truss assembly as a whole.
On the other hand, CDP analyzes one truss at a time and it assumes that truss supports
have zero deflection. When trusses are supported by other trusses, the support at the
connecting point will not have zero deflection. Thus, the CDP is not a good approach
to examine this connection in the assembly. The last observation is the stiff-truss
effect. As expected, stiffer trusses in the assembly attract more load. This effect can
be observed by an increase in CSI values for the stiffer trusses and decrease in CSI
values for the adjacent trusses that are less stiff. Based on this investigation, the
maximum CSI for most trusses in all three assemblies reduces by 6% to 60% because
of 'system effects.'
SDP can help to improve the analysis of truss assemblies by directly including
"system effects" that are not accounted for by the CDP.