In the basement of ¶¶Òõ¶ÌÊÓƵ-BoulderÂ’s College of Engineering and Applied Science, a monstrous centrifuge sporting an 80,000-pound swinging arm and a box to tote hefty payloads whirls a miniature earthen dam at 200 miles per hour.
The scale-model, two-ton dam in the payload box was constructed with a small sinkhole similar to depressions occasionally found in earthen dams around the world. By "flying" the tiny dam at blurring speeds using a 900 horsepower motor, researchers can determine how sinkholes propagate to the surface of full-sized earthen dams under stress, said civil engineering professor Hon-Yim Ko.
The secret is in the swing of the centrifuge, which is similar to filling a bucket of water and swinging it in circles without losing a drop, said Ko. He is testing the dam for the U.S. Bureau of Reclamation.
Capable of spinning 2 tons of material at 200 times gravity, the 400 g-ton machine – among the most powerful centrifuges in the Western world -- uses the force of gravity to simulate the behavior of full-sized earthen and concrete structures under stress.
Chairman of the civil, environmental and architectural engineering department, Ko has conducted a wide variety of experiments in the machine with colleagues and students over the years, ranging from dam safety and the interactions of giant anchors with the sea floor to the behavior of lunar-like soil overlying tubular or spherical structures. In addition to the 400 g-ton centrifuge -- which can simulate the behavior of structures as large as a football stadium -- the department also has a smaller 10 g-ton centrifuge.
The centrifuges are constructed so water can be added to miniature dams in flight to simulate the erosion effects of earthen dam "overtopping" and determine failure points, Ko said. Researchers use "shake tables" that move laterally inside the payloadboxes to simulate soil behavior under earthquake conditions.
Having centrifuges capable of simulating EarthÂ’s tremors is important, said Ko, because soils tend to lose strength and liquefy during earthquakes.
Completed in 1988, the 400 g-ton centrifuge was built by Wyle Laboratories in Huntsville, Ala., and designed by engineer Klaus Cappel. The construction was funded by the National Science Foundation, the U.S. Bureau of Reclamation, the U.S. Department of Defense, Martin Marietta (now Lockheed-Martin), ¶¶Òõ¶ÌÊÓƵ-Boulder and several private corporations. General Electric donated the powerful motor to drive the centrifuge.
"You rarely see dams being torn down -- they either break or are declared unsafe," said Ko. "The safety of existing dams has become an important issue, and our machines are providing a viable way to study it."
Some of the first experiments conducted with the large centrifuge involved testing the strength and failure points of three-foot-high concrete dams reinforced with steel. Coordinated by civil engineering Professor Victor Saouma and funded by the Electrical Power Research Institute, the tests provided new clues that may help mitigate potentially catastrophic incidents.
In real life, concrete dams fail for a variety of reasons, including stresses, construction defects, and the alternating effects of hot sun and cold water. When the centrifuge subjected the tiny concrete prototypes to the same conditions, they cracked, crumbled, split and burst just like the real thing, said Saouma.
"Over the years we have combined our centrifuge work with field studies, lab work and numerical studies to refine the answers to our questions," said Saouma. Most recently he has been testing the tensile strength and fracture toughness of a real dam in Georgia in a project funded by EPRI.
While computer simulations have become a popular technology of predicting the outcomes of geophysical events, physical-modeling studies like those performed in centrifuges and wind tunnels are invaluable, said Ko. "At times, computer simulations only tell you what you already think, and do what you want them to do," he said. "On the other hand, a structure will tell you what it wants to do. That is why these physical experiments are irreplaceable."