Large scale structural testing

I was kind of busy today, so I just saw this fun Boing Boing piece on the Constructed Facilities Laboratory at NC State. The post, written by Maggie Koerth-Baker, consists largely of a series of short videos from a tour of the lab given by its lab manager, Greg Lucier.

The facility seems very nice, similar in many ways to structural testing labs I’ve visited at Lehigh and at my own alma mater. Mr. Lucier does a decent job of explaining things, but I can’t help but think that most of what he said was lost on the members of his tour group. In my experience, even people who have an interest in science and read popular science books don’t understand the first thing about structural engineering.

Part of this, I’m sure, is because structural engineering is all around us; it’s so commonplace that its fundamentals aren’t taught, even in elementary science classes. Yes, everyone learns about balancing forces, an essential part of structural analysis, but that seems too simple and dull, so the topic quickly changes to kinetics so you can do those wonderfully practical ballistics-in-a-vacuum calculations.

Even the basic vocabulary isn’t taught. In this part of the tour, poor Greg is explaining how they monitor the response of a large pipe to a bending load, and he points out the gauges bonded to the wall of the pipe for measuring strain. I doubt anyone in the party understands what strain is, so the value of the monitoring is lost on them.

Elsewhere in the tour, Greg explains their shake table, used to simulate earthquake loading. He gets a question about whether the table can be used to test against any earthquake “level,” and he’s clearly taken aback.

He reluctantly throws out the word “magnitude,” and gets a nod of recognition from the questioner, but you can tell he’s just disgusted with the idea that his cool shake table, which can reproduce the fine details of any ground acceleration record ever recorded, is somehow controlled by one stupid dial that gets turned to some number between 3 and 10. This fixation on earthquake magnitude—which I blame on those bastard seismologists and their fellow travelers in the so-called mainstream media—prevents people from understanding what’s really important in determining how a structure reacts to an earthquake.

(My house, by the way, survived a magnitude 9.0 earthquake last year. Yes, the earthquake was off the coast of Japan and my house is in Naperville. Does that matter?)

The one thing that bothered me about the piece—and this is Maggie’s fault, not Greg’s—is the impression it leaves that structural engineers need to test everything before it gets built.

Buildings, roads and bridges are all designed with a buffer of safety—basically, engineers round up on the numbers, a lot, and design these things to be far more sturdy than they actually have to be. But to make those decisions, they first have to know the physical limits of the materials they’re working with. The best way to do that: Take a scaled version of a girder, pillar, or concrete slab and push it past the breaking point.

In fact, structural engineers are probably less reliant on testing than any other type of engineer. Yes, we need to know the strength of the steel, concrete, or wood used in our structures, and that’s done through testing of small samples, but once we have those fundamental material properties, we can figure out through rational analysis how to size all the beams and columns of a skyscraper. Most practicing structural engineers never perform the kind of large-scale component and system testing done at the Constructed Facilities Lab.

You wouldn’t want to buy an iPhone that hadn’t been rigorously tested before being put on the market. You wouldn’t drive a type of car that hadn’t been crash-tested or put through its paces on a test track. But you routinely walk into buildings and drive across bridges that went straight from blueprint to reality. And it’s perfectly safe to do so.