Systems on the edge of failure sound funny. Ask the mechanic listening for the death-rattle of a dying U-joint or the doctor thumping a patient’s back.
And then there’s the deadly “click” of laminated surfaces coming unglued. Civil engineers have long used acoustic tests to find places where bridge and highway decks are delaminating, getting ready to flake off and fall away from their under layers. These test methods have been relatively slow, though, relying on carefully placed hammer blows, dropping steel ball bearings, or dragging heavy chains. The testers crawl down the road thump by painstaking thump at a top speed of less than two miles per hour.
Researchers at electronics and civil engineering departments of Brigham Young University have developed an elegantly simplified method for pounding the pavement, using the percussion of water droplets to find the weak spots.
They built a simple device (described in Non-Destructive Testing and Evaluation International) from off-the-shelf components: a water tank, a couple of lengths of 6-mm ID flexible tubing, a pump, a flow meter, a good array microphone, an acoustic sampling unit, and a reconfigurable I/O controller. The test rig dripped the droplets two meters onto a concrete floor (with a known weak spot from an earlier repair that didn’t quite take), and picked up the echoes from the impact with the microphone. (The microphone itself was raised well above the splash zone; a tubing “ear trumpet” put its ear to the ground in the splash zone and carried the noise back.)
Brian Mazzeo (he’s the one pouring water on the cracked floor in the photo), Anjali Patil, and W. Spencer Guthrie tested two droplet streams: a 22.5 milliliter-per-minute flow of 7-mm-diameter drops (weighing roughly 1/6 grams apiece) at about 2.3 drops per second, and a faster 94 milliliter-per-minute stream of smaller 3 mm droplets (just 1/70 grams each) at a “white noise” clatter of 110 or so impacts per second. The data—acoustic spectrograms of frequency over time—show clear, lower-frequency ghosting of echoes from buried structural discontinuities.
By adjusting the flow rate and drop volume, civil engineers may well be able to tune the device to reveal specific types of potential failures well before they occur. The method offers inexpensive components, easier clean-up (no BBs or buckshot to chase all over the roadway), and easier maintenance (no jammed hammers to re-set, no worn pivots to maintain).
There are a host of items to work out—finding the right drop-size and flow rates, of course, and coping with real-world wind and sub-freezing temperatures, just for starters—but it’s still fun to see something simple and useful improved by something even simpler.
The objective is to be able to drive a water-spraying test rig across a bridge or over a stretch of road at 25 or 30 miles per hour to detect structural flaws without long lane closures and traffic delays. And beyond that, water-impact acoustic testing could be expanded to quickly and cleanly detect spreading lesions in all sorts of materials, from airplane parts to fiberglass boat hulls.
Images: Brigham Young University