like Calvin from “Calvin and Hobbes” asked How do they know the load limit on bridges?

Question

Inspired by Calvin’s same question where his Dad answers “They drive bigger and bigger trucks over the bridge until it breaks. Then they weigh the last truck and rebuild the bridge,” how do engineers determine the actual load limit on bridges in the real world?

In: Engineering

Answer 1

Materials:

The materials you use to make bridges all have known strengths. The steel is rolled in a mill and sampled for testing, and is reliably very close to spec, often a modest amount above to ensure statistical confidence that even the “dud” parts of any production run meet spec. The concrete strength is estimated based on extensive experience with the mix designs (ratios of cement, sand, gravel, water, and the specific properties of the sand and gravel available to the manufacturer). This is a more variable material than steel, so each batch is tested at placement to ensure it meets the spec properties. If you’re using aluminum and wood the same deal applies.

Engineering:

You use engineering analysis to determine how the loads on the bridge translate into forces in the structural components, then you make sure the structural components are stronger than the demand on them. There are also design checks for deflection and how much the bridge vibrates and so on. 

This has been a known process for over a century, so covers most bridges. With older bridges a retroactive analysis can be applied along the same principles, often taking small material samples so you know the strength of it all.

 The final load rating also includes a bunch of safety factors, adding up to a factor of safety of about 2 on the theoretical design load, and about 3-4 on the actual highest load the bridge will see. This is for a few reasons, but part of it is to give some slack to cover gaps in maintenance, construction errors that might not be caught in a difficult environment to inspect, and because almost every bridge is a prototype – it’s not like designing a jet engine where you build a bunch of test engines and test them to destruction. You build *one* and put it into service. With virtually every one being unique sometimes something in the design throws a curveball so it’s nice to have some spare capacity.