I built this bridge in 7 hours one Sunday. I did this mostly for fun, but I learned a lot about bridge design (see the "Final Thoughts" section for comments). The span length is about 74 inches and the height is 30 inches.

Unlike my previous bridges ( 1 and 2 ), this bridge is much lighter in construction and it makes more efficient use of the materials. The previous bridges could withstand loads greater than 30 pounds which is far more than necessary. This new bridge is better suited for a roadway for small plastic vehicles or for trains. It is 16 bricks in width so it could easily accommodate two sets of tracks.

See extra pics here.

The arch is made of 7 segments, each about 14.5 inches long. Each segment is joined at an angle of 22.5 degrees and uses #3 Technic axle connectors. All segments of the arch are connected at the same angle so the arch curve is circular (as opposed to an elliptical curve).

Each segment of the arch is a simple truss made of Technic beams connected with a lattice work of axles and connectors. Actually, it's not a true truss as the lattice diagonals don't connect at common points. The result of this construction is that significant bending moment is created locally in the main members (the Technic beams). However, the Technic beams are relatively strong and the resulting structure is quite sound. Whatever bending moment generated is insignificant.

The principle of this type of construction is that a roadway, or railway, is built on top of an arch that is put into compression. If the ends are restrained from spreading apart, the arch is quite strong and capable of supporting the roadway on top.

This bridge did spread apart quite easily! When loaded, the ends went scooting off therefore spreading the arch apart. I guess in real life, the bridge abutments are firmly anchored to the ground and resist movement. For my model, I can always weight down the foundations or place them against something (e.g. a wall).

I tried a much different approach for the cross bracing as compared to my previous bridges. Because this bridge will carry relatively light loads, I felt that axles would be sufficient for the bracing. I connected 8L and 10L axles with connectors and then snapped them into place diagonally between the main members of the bridge. Although these long axle members cannot carry much tension, they cannot carry much compression either (they're too long- buckles easily). By using a lot of these axle diagonal members, the result is a cross-bracing system that is strong enough.

This bridge is presented in an unfinished state. I have yet to add some sort of roadway or railway and the cross bracing isn't complete, but I probably won't do anything more to it. It's not perfect and I'd prefer to rebuild it if I wanted a permanent structure.

Building this bridge was a good learning exercise. Here are a few major points:

1. The arch should be more stiff and resistant to bending. The connection between the segments could be made stronger.

The arch is normally in compression. However, given that this is only a model, the bridge will be picked up and moved and odd loads will be placed upon it. People will like to touch it and test its strength by pushing down on it. The foundations may not always be fixed and will be susceptible to sliding out.

Given this, the arch will also perform as a curved beam. It will see significant bending moments as well as the normal compression. Also, the roadway is supported at finite lengths so localized bending will occur (from loads applied to the roadway) in the arch.

2. The overal dimensions should be calculated and used to set up the bridge.

Because the arch is flexible, the ends can be easily moved in and out making the span longer or shorter. The ideal position is hard to attain because of this.

By using a little math, the perfect (or ideal) dimensions can be calculated. These ideal dimensions represent the unloaded, undeflected, and unstressed bridge. When setting the bridge up, the exact distance between the ends (which has been calculated) can be measured out on the floor and then the bridge ends can be placed at the correct distance.

Also, by calculating the bridge dimensions, the exact length of the members can determined prior to building them. I calculated the lengths of the vertical members under the roadway, and the horizontal members of the roadway. I then built them to the calculated distances and the result is a perfectly horizontal roadway (assuming the ends are set at the correct distance apart).

3. For efficiency and better appearance, the roadway should be supported at regular intervals.

As you can see from the pictures, the vertical members that extend from the arch to the roadway are connected only at the joints between the segments of the arch. The result is that the horizontal roadway is supported at uneven span lengths.

The vertical members can be easily connected at any point along the arch. They can be placed so that the roadway on top spans evenly between vertical supports. This means each roadway section is identical and uniform.

Also, if you're building the arch from straight sections (like I did), the middle section can be horizontal (like my bridge) and then used as part of the roadway. This eliminates the need for a middle roadway section, as the roadway (or railway ;-) can be built right on the arch structure itself.

See extra pics here.