Authors: Bradley Peniston
Haggett held a press conference that afternoon to describe the repair plan. The concept could be simply stated: cut out a big chunk of the ship, build a replacement, and weld it all back together. But this belied the actual complexity of one of the most difficult repair jobs an American shipyard had ever attempted. The
Stark
had been repaired with modular techniques at Ingalls Shipyard for $90 millionâmost of which went to replace the CIC and its combat systemsâbut its keel had not been broken, and the module was lowered onto the damaged ship from above.
33
The
Roberts
repair would require BIW to shoehorn the replacement sections into the bottom of a completed frigate. The job had to be done with great precisionâin some places, less than one two-hundredths of an inch. A misaligned hull section could slow the ship, reduce its gas mileage, and even make it noisier in the waterâall of which would make
Roberts
a poorer warship and a better target for prowling submarines.
That such a repair was possible at all was due in large part to the ship's modular construction. BIW had delivered its twenty-fourth and final
Perry
frigate about a year previously, bringing to a close one of the century-old yard's finest chapters. But it still had the plans for each of the thirty-six massive chunks that made up each frigate, and the tapes that guided the cutting machines that shaped their steel plates, and workers who could all but turn them out in their sleep.
But the blast had mangled two of the modules, which forced a difficult choice: should they be replaced individually or with one giant piece? After much discussion, BIW engineers decided that a single chunk would be easier to work with. But that module would be a monster: 310 tons of gas turbines, generators, bulkheads, stringers, hull plates, and all the pipes, wires, and gear that make up the interior of a warship. It would be, by far, the largest single chunk of a ship the yard had ever handled.
Work had already started, Haggett told the assembled reporters. The semiautomated modular construction had allowed the Bath workers to jump right in. The tapes that drove the automated cutters had been broken out of storage, and fabrication had begun in BIW's Hardings plant in early June, within a week of the contract award. The pieces would be assembled in Bath and then barged south to Portland.
Haggett said the
Roberts
repair would be a “much more challenging job” than BIW's World War II repairs. It would be “an engineering experiment in a sense,” but a “workable” task, he said. Ultimately, he estimated, it would involve up to five hundred workersâabout 5 percent of BIW's workforceâand would likely cost about $96 million, split roughly evenly between parts and labor. The job would be done around the end of 1989.
Winn Price, a BIW program manager who would run the repair effort, told reporters that the ambitious schedule reflected “an intense amount of pride in this ship, and an intense desire to get her back to sea.”
34
THE DAY AFTER
the press conference, the frigate USS
Vandegrift
(FFG 48) arrived in Kuwaiti waters in company with the supertanker
Townsend
. It was the sixty-seventh escort mission of the year, the eighty-ninth of Operation Earnest Willâand the last.
Thanks in part to the mining of the
Roberts
and its violent aftermath, Iran and Iraq had agreed to a cease-fire in August, and on 26 September 1988 the White House ended convoy operations in the Persian Gulf. Since July 1987, U.S. Navy warships had escorted 252 vessels through the Gulf. None except the
Bridgeton
was damaged by enemy action while under way.
35
THE
ROBERTS
ENTERED
Portland's dry dock once again on 6 October. Support blocks had been built parallel to the dry dock's keel, but about thirty feet off center, leaving room for the replacement module to be placed next to the ship. Computing the proper settings for the ballast tanks was a relatively straightforward matter for a dry dock that had been built to handle several ships at once. And unlike the Dubai dockmasters, BIW's engineers had gotten the chance to measure the hull's deformed contours. They shaped the blocks to allow for the one-degree bend in the main deck, and consequently, the
Roberts
settled down without further injury.
When inspections had confirmed that the frigate was safely ensconced, the first order of business was getting rid of the sand ballast that had come over from Saudi Arabia. A pair of holes was cut in the temporary engine room patches. But when local environmental authorities got a look at the grayish-white sandbags with the Arabic script, everything came to a quick halt. Just what had been the purpose of those bags? Concerns grew when someone managed to translate the word “flour” on the plastic webbing. Sealed samples of the sand were flown to the U.S. Department of Agriculture in Washington. Bill Forster, who ran the department's quarantine section, worried that the bags might contain khapra beetles, among the world's most voracious consumers of seeds and grain. But an inspection showed the bags and their contents to be free of insects, fungi, and disease, and workers resumed clearing the engine rooms of the sand, slitting open the bags and sucking them clean with vacuum hoses. The job took a week.
36
The next step was more ticklish: removing the one-degree bend in the ship. Careful measurements had shown the stern to be down precisely thirty inches. Bath engineers had determined that it could be raised, and the ship returned to its normal configuration, by a phalanx of jacks pressing upâcarefully, carefully!âon the aft section of the hull. “It's not a usual operation for a shipyard,” said Dave Ward, who coordinated the jacking effort for BIW.
Preparations began by welding stools to the hull. These small platformsâfive pairs of themâwould give the jacks something to press against and would distribute the force to the ribs inside the hull plates. This would allow the jacks to lift the hull without punching holes in it. Each jacking stool was attached along a frame, spanning three longitudinal stringers. There were also nine more cribbing stools, which would help bear the burden of the hull.
Next, Ward's team removed the blocks aft of frame 280, replacing them one by one with towers of steel and concrete, one for each stool. Each tower was topped with four slim wedges of oak laid flat, then sledge-hammered into position to assure a tight fit against the stool.
Each of the ten jacking towers sported a pair of orange-painted aluminum cylinders. About double the size of a Quaker Oats canister, these jacks would lift one hundred tons when energized by a 10,000-psi pump.
Common practice called for jacks to be pressed to no more than three-quarters of their rated strength, so this setup provided fifteen hundred tons of lifting power for a thousand-ton job.
37
Before the jacking rig went under the ship, BIW workers assembled it ashore and tested it to capacity. They also brought in extra pumps to ensure that once the jacking started, nothing would slow it down. A surveyor's transit was set up in the temporary engine room, its barrel trained directly down the tube that would hold the propeller shaftâor rather, trained along the line that should go straight down the tube. As the jacking proceeded, this transit would tell Ward and his team when they had the ship back in its original shape. Finally, Ward's workers sliced open the superstructure from roof to main deck. The thin V-shaped gap at frame 280 widened from the width of a saw blade at the main deck to about five inches at the roof. This would allow the superstructure's aft end to rotate one degree as the stern came up. They made a second vertical cut in the hull at frame 277, slicing through plating, stiffener, pipes, and wires, for the same reason. The cuts were offset and stiffeners added to make sure the main deck unbent, and did not further distort, at frame 280. This rotation was no minor matter, for it made the operation much trickier than a simple vertical lift. The stool at frame 286, for example, would rise only two inches during the jacking, but would move aft about twice as much.
The jacking began early one wintry morning. Workers with hardhats and walkie-talkies stationed themselves all along the jacking stations. Pumps hummed as the four aft jacks were pressed to about half their capacity. Next, the six forward jacks received hydraulic charges. All but imperceptibly, the hull that towered about the workers' heads began to move. Men with sledgehammers pounded the oaken marrying wedges toward each other, taking up the slack, keeping the jacks firmly pressed against the stools. Ward had his team continue jacking until one of the stations could accept a four-inch oak plank. The marrying wedges were loosened, the board was inserted, and the jacking began anew. From time to time the jacks were pressurized in alternate sequences as required by the tilting or rotation of the hull. Over time, layers of four-inch planks were replaced by foot-thick timbers, and still the work went on, centimeter by centimeter, all ten jacking crews working to keep their station working in unison with the others.
The sun rose above the islands in Casco Bay and traveled across the sky as the painstaking process went on, and the light was dying behind the waterfront warehouses when the transit finally indicated success. It had taken twelve hours to undo a bend that a mine had caused in microseconds.
The following morning, Ward and his crew rechecked the alignment. Custom wedges were cut and installed on the cribbing towers, and the jacks were depressurized and released.
In the following weeks, the temporary hull module built by Dubai Drydocks was cut from the frigate and lowered to the pontoon deckâthe dock's “floor.” The skeg, damaged in the post-blast docking, was replaced in prefabricated sections. Nearly thirteen hundred square feet of warped steel plating was removed from the superstructure and several longitudinal frames repaired; new plating was welded back on.
In Bath, workers wrapped up the structural part of the replacement module. The 160-ton assembly of framing, stringers, and hull plates, looking rather like a bathtub squared off on both ends, was rolled out of the long main assembly building, across the concrete apron, and into the paint shack, where it was sandblasted clean and freshly painted. So far, the module contained little of the machinery it was built to carry. The reason: the Japanese-designed level-luffing crane could handle no more than 200 tons. Even the Bath engineers weren't sure just how heavy the replacement module would be when completed, but they were guessing 350 tons.
So the candy-striped crane lifted the incomplete structure onto a 200-by-56-foot barge brought up the Kennebec River for the purpose and then added two gas turbines, a reduction gear, and the other large chunks of machinery. The large components were secured for the trip down to Portland, but final adjustment would wait until the module had been welded into the ship. High winds and November seas delayed the departure for two days; the barge finally left on its forty-mile trip down the river and the coast.
38
But there remained a quandary: how would this heavy moduleâit had ultimately weighed in at 315 tonsâget from barge to floating dock? BIW dockmasters considered ballasting the barge until its deck was level with the pontoon deck and dragging the module across the gap. They
also mulled sinking the dry dock to let the barge come aboard and then lowering the module to the pontoon deck with jacks. Both options were eventually deemed too risky. Instead, BIW sent all the way to Virginia for a floating crane called the Chesapeake 1000, a herculean piece of equipment named for its ability to hoist a thousand-ton load. It arrived on its own barge and was so tall it required FAA approval for use near Portland's airport. BIW bought one-hundred-ton wire ropes for the lift and rigged them from the Chesapeake's hook through pairs of sixty-ton chain-falls to four points on the module.
39
The lift operator pulled a lever, and the module rose slowly off its barge. BIW workers adjusted the chainfalls so that the module hung level in the air, sixteen feet above the calm harbor waters. A tug nudged the crane's barge alongside the dry dock so that its long boom extended over the dock's six-story wall, the heavy module dangling over the pontoon deck. The dry-dock wall prevented the crane operator from seeing his load, so a lift coordinator walked along the wall, relaying instructions via walkie-talkie.
The tug pushed the barge some 240 feet along the dry dock, until the module was next to the clean-cut hole in the
Roberts
. Three rails had been set on the pontoon deck, perpendicular to the frigate's centerline. Working blind, with only the radio instructions to guide him, the crane operator lowered the module to within one inch of the rails. BIW workers adjusted its position with cables and ratchet come-alongs and lowered it the thumb's width to the rails. When they measured their work, they found that the 315-ton chunk of ship was within a quarter inch of the ideal position.
The final leg of the module's journeyâshoving the module into its hole and jacking it up into positionâwas a sufficiently complex endeavor that BIW sought outside help. The company hired Atlantic Industrial Contractors of Richmond, Virginia, who had set up the system of rails and jacks that now supported the module. In a single daytime shift, Atlantic workers used the jacks to move the module on graphite-coated rails, easing it sixteen feet horizontally until it was directly under the ship.
The final lift resembled the stern-jacking operation that had unbent the ship. Over the course of a day, BIW workers used eight one-hundred-ton jacks to lift the moduleâand the skidding rails and jacksâstraight up,
pounding in oaken wedges and piling up boards until the steel form fit snugly in its void.
The final adjustments were measured in hundredths of an inch. The path of the propeller shaft had to be straightâreally straight. BIW workers had already fixed a fingernail-thin dent in the shaft caused by the explosion. (It would have been preferable to simply replace the shaft, but no extras had been manufactured for the
Perry
class.) A laser transit was trained up the shaft line, and with infinitesimal adjustment, the module was aligned for the final time. Welders wielded their tools, and the shattered ship was whole again.