Dynamic Deeds Build Clemson's Unique Test Facility
After braving a "perfect storm" of construction challenges—including mucky conditions and seismic- and wind-load concerns—the builders behind Clemson University's first-of-its-kind wind-turbine drivetrain testing facility in North Charleston, S.C., have finally found safe harbor as the university prepares to begin commissioning.
"We had heavy loads on muck in a seismic area with flooding potential and high wind loads due to hurricanes on a brownfield site," says James Tuten, program manager for the Clemson University Restoration Institute (CURI), the project owner. Indeed, project officials would have been hard-pressed to find a more challenging site for constructing the roughly $100-million testing facility, which was financed in part by a $47-million Dept. of Energy grant using stimulus funds.
Adding to the challenge was the need to build the state-of-the-art facility—housing some of the world's largest equipment of its kind—within an existing structure, as required as part of CURI's sustainability mission of revitalizing historic and urban structures.
To retrofit the 1980s-era Navy warehouse to today's codes for wind and seismic loads, contractors first redid the existing building foundation. They installed 432 steel H-piles, ranging from 46 ft to 57 ft in length, and placed 980 cu yd of concrete. To beef up the vertical structure, crews added 1,200 tons of steel columns and beams.
Keeping it Steady
As a platform for testing drivetrain equipment, the heart of the center revolves around its two test-rig systems, which are designed to deliver 20-year-life loads to drivetrains within a roughly six-month span. Essentially, each of its two test beds is a "big vibrating piece of equipment," Tuten says, with each generating forces in an unsteady, cyclic fashion.
Moreover, the two test structures are among the most powerful in the world. The smaller of the two is driven by a 7.5-MW gearbox, while the larger features a 15-MW, 341-ton gearbox that is considered to be the world's largest. Designed as secured, independent systems, the rigs can simultaneously accommodate confidential drivetrain testing by competing manufacturers.
Whether working separately or simultaneously, the dynamos will produce "tremendous" dynamic loads, says Tuten. The two separate bays for these systems—measuring 26 ft by 86 ft and 50 ft by 100 ft, respectively—feature dynamic, independent foundations.
The engineer of record, AEC Engineering, opted for a complex system utilizing friction piles, which project officials believe is a first for this type of facility.
"We are unaware of anybody else who has ever built a dynamic piece of equipment on friction piles," says Tuten. He adds that the approach required extra analysis. "We couldn't let [the testbed] become a vibrator and watch it just vibrate its way down."
As independent, isolated structures, the test-bed foundations had to be able to avoid transferring any vibrations, explains Thomas Lorentz, senior vice president with AEC Engineering.
"The foundations required isolation from the existing structure such that external vibrations were not induced into the test specimen and that test vibrations were not transmitted to the facility structure," he says.
With the combined concerns of load and mucky underground conditions, the support structures would need to be especially deep, with the smaller requiring a 10-ft-deep concrete foundation and the other needing a 13-ft-thick base.
Supporting the 7.5-MW section are 40 35-ft-long steel shoring piles, 88 70-ft-long steel piles and an estimated 250 tons of reinforcement steel, ranging in size from No. 4 bar to No. 11. For the larger test rig, 54 35-ft-long steel piles and 115 75-ft-long concrete piles form the base, along with 650 tons of reinforcement steel.
To enable flexible positioning of the equipment under test, AEC included an anchor sleeve system that allows for a range of load locations for the various drivetrains, Lorentz adds.
In each case, however, the piles' depth—and location within an existing structure—required some creativity, says Chris Palmer, project manager for Choate Construction Co., the general contractor.
"We had to modify the pile-driving rig to drive the piles 20 ft at a time," Palmer said via e-mail. "We would then have to full pin moment weld the next 20-ft section, which took about three hours per weld." In all, he estimates it required more than 1,200 worker-hours just to weld piles back together.
Altogether, the two test rig beds utilize 900 tons of rebar for their respective 10-ft- and 13-ft-deep concrete foundations. Most of the test-rig foundation rebar was No. 11 bar, measuring 60 ft in length, Palmer says. At that scale, placement would require as many as eight workers hoisting each bar.
For the concrete placement, only self-consolidating concrete would work in this situation, he adds.
"With the amount of rebar we had in each foundation, it would prove impossible to get traditional concrete and vibrators down 13 ft and know you have a solid foundation," says Palmer. Even so, designers had to be convinced it would work.
Key to that sales pitch was the recent experience of local concrete contractor Cooper River Contracting, which had used the method on another area project.
"They were really the driver of that [choice]," says Palmer. "That was a huge asset to this project as far as time and constructibility of the foundations. I don't think we would've gotten [the foundations] built" otherwise. The 7.5-MW section required 880 cu yd of concrete, while the larger one consumed more than 3,600 cu yd.
Precision Planning
Clemson's Tuten says he's most impressed with the accuracy with which these support structures were constructed, as the tolerance of most components was 30/1,000th of an inch, he says.
"We knew QA/QC was going to be critical," says Lorentz, characterizing the complex reinforcing system as "quite the jigsaw to put together." After AEC and Choate 3D-modeled everything, onsite inspection and metrology became key. For instance, the anchor sleeve system, which was prefabricated, was laser scanned a total of four times prior to concrete placement: at fabrication; delivery; installation; and for as-builts.
"There were hundreds of individual components that went into building each foundation, all of which had to relate back to each other to within 1/16 of an inch in every dimension, X, Y and Z," Palmer says, adding that failing to meet tolerances would result in the drivetrains not aligning.
Construction of the 15-MW test rig's load application unit (LAU) structure—the vertical structure that houses the gearbox and drivetrains—was another feat, says Tuten. For instance, after the initial positioning of the galvanized-steel load disc, but before concrete placement, the massive prefabricated ring "looked a little below datum" to him. But that was by design.
"We had to anticipate what the theoretical deflections are during construction [versus] during service," explains Lorentz. "What are the deflections going to be from a dead load? And how do we get it back to a relief position and a zero position at end of construction?"
Installing the 15-MW gearbox posed the final hurdle. Once it arrived at the Port of Charleston, it took roughly one month to get the item in place. Tuten explains it this way: "A 341-ton gearbox does not go anywhere very fast, and if it does you're in real trouble."
To accommodate its installation, the contractors left a roughly 20-ft-by-20-ft portion of the structure unbuilt. That meant that installation of such items as flooring and utilities that support that eastern-most portion of the building couldn't begin until the gearbox was in place.
The storm of construction now over, Clemson is preparing to start commissioning the test rigs, a process that should finish later this year.
