Analysis for ME 14
Bridge Competition (Learning from Failure)
Jefferson Cam and Xinghao Huang

The second midterm of ME 14 statics was the bridge competition. Two people would form a team and use balsawood to build a bridge that was 40 cm long, 10 cm wide, and 21 cm tall (maximum dimension). The bridge also had a 25 cm gap between its two main supports and 10 cm clearance.  Only Elmer's glue was allowed to use at the joint, and no spacer allowed to enhance joint. ​The team that had the highest load to weight ratio would win.

The 40-hour effort input didn't make us win, but learn a lesson...
I and my roommate Jeff Cam formed the team and built our bridge.  We used a software developed by Pre-Engineering Inc. called ModelSmart3D, which can simulate balsawood in different structures. Below are two designs we made and their simulation at the breaking force.  
Picture
The Triangular Design
Picture
The Arc Design
We eventually chose the triangular shaped design. We chose it because:
  1. It's easy to build since the arc one had two many angles.
  2. More load. The triangle shape can distribute force from top to bottom.
  3. Light. The final weight was only 34.8g. 

We started building its sections. We printed out the cross sections of the bridge on paper and and used it as a reference template to build the side first.

We then built the two triangle bases.
Picture
Jeff is measuring the dimension
Picture
Paper template
The two sides were connected by horizontal beams. When gluing, we had to lie it down on its side and put load on it.
Finished: 

Failure 

On the test day, Professor Kirk used a compression actuator to put a load on our bridge. The peak load would be recorded to calculate our final score. 

When it was our turn, our bridge only hold around 50 lbs of load and only last 5 seconds. But before that, we got 1080 newton in our simulation. Why the simulation is so different from the actual situation?
There are three reasons we failed: 
  1. The simulation cannot simulate the real situation. It can only provide a reference for our bridge. ModelSmart3D assumes all joints are rigid and the breaking point only happens on members. However, the glue we used was very weak, so the joint broke before the force reached the breaking point of balsawood.
  2. As shown at the red circle on the left, we didn't put these two triangular members at the correct position. When force went down from the top, the triangular members send force to two sides, but they need a structure to prevent them from moving toward the side. It would be better if we made it contact the right side of the horizontal member instead of on its top.
  3. If we do the second point correctly, we also need to prevent the horizontal member from moving toward side. So the bottom triangular members should cover the two ends of horizontal member. We shouldn't rely on glue since its property is not as stable as wood members.
So, never trust simulation and glue. Trust the structure. It would be better if we can build one more prototype and test it.