Building a Bridge - Engineering Part 2

 I've worked on, or built, a few bridges during my career. Most were beam and girder, with one being a large cable stay bridge. When looking at the design, the fact they were built is a testament to wonders of modern engineering. The fact they stand for so long shows the design wasn't a foolish whim of a demented person, and after working on some older structures, the fine tuning of the design of newer bridges indicates some designs are now obsolete; even though the less desirable designs are still in use and safe. 

Not all bridges make it to the construction phase. We had a local bridge designed to the point a three dimensional model was made to scale. A historical society, and government officials, became involved with preserving the existing swing bridge, and the new bridge was cancelled. The old bridge was refurbished at a tremendous cost, was still an old, untenable, high maintenance bridge, and regardless of how ridiculous the effort was, remains as something that taxpayers shouldn't have to pay to maintain. 

The final design of a bridge requires knowledge of materials, the distances to be spanned, and the cost. Steel is one of the best materials, but requires coatings, which can be expensive over time. There is a steel that corrodes on the outside, but will not have deep corrosion. It's substantially more expensive than regular structural steel, but I have seen it used on overpasses. It's uglier than homemade soap, when the rust "bleeds" onto the concrete caps, and columns, supporting the girders made from the steel. Still, steel is the best material for the type of design, and the long spans required for flyovers, and other long spans in urban freeway congestion. 

Steel can be designed to span long distances, where concrete can't without overhead supports. What girders are used depends on the span distance, whether there is a typical wider flange beam that can handle the load, or specially fabricated girders with camber to offset deflection of the weight of the deck. On larger spans, the design can be a multitude of different steel configurations, including trusses, arches, and box beams. Regardless of the bridge design, steel will be used somewhere in the construction.

Concrete usually requires less maintenance, but concrete has limitations. The biggest limitation is how brittle it is. If a span has nothing but concrete, it will stay together for awhile, but the expansion and contraction will eventually cause large sections to crack, and fail. That's where reinforcing steel is necessary, and planned joints needed for temperature changes in the concrete. 

The "rebar" holds the concrete together, and distributes the load evenly. The more rebar, the more strength. This allows some of a bridge decks to be as thin as 8 inches, and still handle the weight of a full loaded semi at highway speeds. Concrete can be used for long span girders, but the only application I've seen is box girders with substantial dowels, or stay cables. What makes this design convenient is that huge sections can be precast, hauled to the site, and placed by cranes. Still, the concrete has to be reinforced, and maintenance requires constant monitoring of the concrete for corrosion of the reinforcement. Concrete always has cracks, and water will migrate to, and along, the reinforcement steel. Over time, the rust will expand, concrete will spall, and eventually a major section will fail.

For shorter spans, concrete girders can be used. These are prestressed before the concrete is poured with steel cable tendons. Camber is also cast in the girders. The camber required for the deck deflection is predetermined by years of testing, and even with the best of designs, the deflection may not be as wanted. I've driven over bridges that have a "lope" between spans, where the girders didn't deflect as designed. On one bridge, the inspector was constantly harassed about "his" bridge, which had this problem. Subsequent examination showed all elevations before the deck pour were accurate. The girders just didn't deflect as designed. 

With the design finally determined, the next phase is plans, specifications, final anticipated costs, and the approval of the highest powers involved. This can be very time consuming, but if many standard designs are involved, and the project is relatively small, the time can be very short. That requires another post.