GEEN 2851 - Truss Project

MATERIALS TESTING

Before we could begin designing the truss, we had to test the strength of the pine wood and the wood glue. We used a four-point bending test to determine how much bending stress the wood could bear before it failed. We placed a wood sample of known dimensions between four rollers in the universal testing machine, and applied force until the beam failed. We centered our 12” sample along its long edge on two support rollers 11” apart, which left a 0.5” overhang on each end. We then placed two load-applying rollers on the upper part of the wood 4” inside of their respective lower rollers. We also conducted shear tests to determine the stregth of the wood glue for our truss and the strongest combination of wood cut and gusset grain direction. For the test, we built 3 different pieces, each with two pine beams and two pine gussets. Each piece had different combination of grain direction for the pine beam (quarter sawn or flat sawn) and the gussets (parallel grain or perpendicular grain) and was connected with Titebond III premium wood glue. 

Once we determined gathered our test results and data, we compiled it into a spreadsheet with the data from the other groups in class. Then, we determined the mean value for max bending and shear that the wood could withstand before failure, and we determined the standard deviation for each of these values. Once we knew these numbers, we were able to move on to designing the truss.

TRUSS DESIGN PROCESS

After discovering the stregth of our wood and glue, and the optimum gusset orientation and wood grain pair, we began to research types of trusses we could build. We decided to build a Howe truss because it seemed straightforward to manufacture and they are designed to withstand the exact loading conditions we tested our wood with. Once we determined our truss type, we began drafting possible models using Onshape and conducting analysis to estimate what load our truss would fail at. We first used the joint method to analyze the force that would be in each member under an 800 lb load applied at the top center joint for our initial design, then used Matlab and Excel to iterate our design and determine the force in each member under the applied load. We also used Excel and Matlab to determine the factor of safety and strength to weight ratio of our truss. Our second to last design (see Fig. 4 below) included a zero force member, which we opted to remove (see Fig. 5 below) in order to improve our strength to weight ratio. 

FINAL TESTING

In preparation for testing our truss, we calculated the load we expected our truss to fail at to be 2300 lbf and a strength to weight ratio of 2560 lbf/lb.  However, when we finally tested our truss, it failed at 1294 lbf, at a deflection of about 0.212 inches (see Fig. 6 below). This load was well below what we expected. Our truss failed along a preexisting crack in the wood that appeared during construction, which we believed to be a minor fracture that would not affect the truss much. Because of this, we did not account for the flaw in the wood when calculating our expected load. 

TECHNICAL REPORT FOR GEEN 2851 TRUSS PROJECT

To summarize our process from materials testing to the final truss testing, our team wrote a technical report which dives into the details of our project. In the report, we go into depth about the results from all our materials testing and how we used them in our design process. We also explain how we calculated our failure estimate for the truss, and our strength to weight ratio.