A USAir maintenance official had realized that the plane would be ideal for a series of destructive tests of some of the most bizarre theories—that a rudder cable had snapped in flight or that a fat guy had stepped through the floor onto the cable. No one had pushed too hard for them. They were the kind of tests that might bend an airframe, so they couldn’t be performed on a jet that would ever be flown again. But the retired plane gave them a perfect opportunity. The tests would allow the investigators to see the forest instead of the trees. They had run hundreds of experiments on individual components of the 737. Now they could see how everything on the plane worked together.
On the first day of the rudder tests, Haueter and Hall held a press conference at the Museum of Flight, which was adjacent to Boeing Field. The podium was set up directly beneath a green-and-white replica of the B&W, Boeing’s first plane. Haueter played a video and explained how the tests would be conducted. Hall then announced that he was appointing a panel of “the greatest minds in hydraulics” to review the NTSB’s work on the Hopewell and Colorado Springs crashes to see if there was anything else they should do. It was an extraordinary step that showed the NTSB was practically desperate.
Haueter didn’t like the idea of an expert panel. Why did they need a bunch of so-called experts? Phillips knew more about the 737 rudder system than just about anybody on the planet. It was as though Flail was saying the NTSB wasn’t smart enough to figure it out. But Phillips had actually been one of the people behind the idea. He and another NTSB engineer thought it would be helpful to have an impartial panel to bring new brainpower and validate the safety board’s work. Maybe fresh eyes would see something that the safety board had not. Hall said his direction to the panel would be this: “If there’s anything we missed, tell us.”
As Hall spoke in the museum, the rudder tests on Ship 213 were under way down the street in a hangar where Boeing had once fixed B-52s. The pipsqueak 737 was parked between two gargantuan 777s, the shiny new Boeing aircraft that had just started flying. The tiny USAir plane had been opened up like a patient on the operating table. A tail section had been removed so Boeing technicians could attach wires to the PCU. Two Boeing vans were parked beside the plane. One was filled with computers and test equipment that would keep track of the results. The other van was like a life support system, pumping the 737’s systems with false information so the old bird would think it was still flying. A gallon of Starbucks House Blend was parked beside the plane, to provide fuel for the investigators.
The fat guy theory had been discounted by just about everyone. McGrew joked that it was possible only if the fat guy wore high heels, which would have broken through the floor. But everyone had agreed to one last test. The plane’s floorboards had been removed so the investigators could see the cables that ran beneath passengers’ feet. Technicians climbed into the cargo bin, where they could look up through the cables into the passenger cabin. They hooked a ratchet onto the rudder cable that would simulate the weight of the fat guy. By cranking on the ratchet, they could add weight in 50-pound increments. The device had a weight gauge so they could watch as the guy went from skinny to obese. Other investigators stood outside the plane, watching to see if the rudder moved. If it slammed hardover, they would know the fat guy was more than just a joke.
“Ready?” someone asked.
“Okay,” Phillips said.
The guy started at 50 pounds, more of a kindergartner than a fat guy. No movement of the rudder.
He got heavier, up to 100. Still no movement.
At 150 pounds, the rudder barely budged. At 200, there was a slight movement, and at 250, a tiny bit more. The instruments in the van showed the rudder had moved only 3.2 degrees, an insignificant amount. (By contrast, the rudder on Flight 427 actually moved about 21 degrees.)
Later that day they tried the cable-cut test. Someone sat in the cockpit and moved the rudder pedals to make sure they worked properly, doing slow sweeps back and forth. Then the pilot took his feet off the pedals and a technician cut one of the cables with a big bolt cutter.
Twaang! The noise echoed like a gunshot through the huge hangar as the cable snapped and recoiled through the plane like a broken rubber band. Outside, Cox saw the rudder panel shudder, but it didn’t turn. Technicians installed a new cable and cut it in a different spot. Twaang! But again, the rudder didn’t turn.
They did more than one hundred other tests and the plane passed them all. The rudder system still seemed to be invincible. But Phillips remained surprisingly optimistic. He knew they were still a long way from finding the cause, but he was patient. Each test gave them new data and brought them another step closer.
Across the street in the M-Cab simulator, Boeing’s lobbying campaign was under way.
M-Cab had become a tool of persuasion to show how Emmett and Germano could have prevented the crash. At McGrew’s suggestion, Boeing engineers had installed a switch that allowed anyone sitting in the pilots’ seats to take control of Flight 427 and, without much effort, to keep the plane from crashing. It was a roller coaster ride with a message. Point the nose down (Emmett or Germano had mistakenly tried to pull it up), twist the wheel, and you have saved 132 lives.
Boeing invited Hall to ride the latest simulation to see how easily the crash could have been avoided. If Hall, a country lawyer who was not a pilot, could save the plane, surely Emmett and Germano should have.
Boeing test pilot Michael Hewett, a balding former navy pilot who looked strong enough to bench-press a 737, led Hall across the ramp into the white cab and invited him to sit in the right pilot’s seat. Hall would act as the first officer on Flight 427. The simulator would reenact a rudder hardover, and then Hall would have the opportunity to recover the plane to keep it from crashing.
Hewett believed that pilots had to assert themselves in the sky but that the crew of Flight 427 had failed to take charge. He said the cockpit tape showed that Captain Germano had “no command presence.”
In the simulator, Hewett’s point was simple: The pilots could have avoided the crash by easily turning the wheel completely to the right and pushing the stick forward to let the airplane gain a little speed. The 737 would have lost altitude, but everyone would still be alive. Hewett summed it up by saying that to pilots, “speed is life.”
Hall buckled himself in and M-Cab started its imaginary flight, cruising along at 6,000 feet. Cox was also riding in the simulator, sitting in the observer seat just behind them. Hall had invited him along because he knew Hewett was going to do a hard sell and he wanted Cox to provide a counterpoint.
Hewett started by demonstrating the recovery himself. He said the standard reaction time for a pilot allows three seconds, so he would count off before he recovered. When the plane started to go nose down, as the phantom pilots pulled back on the stick, Hewett flipped the switch to take control of the plane and counted, “One-thousand-one, one-thousand-two, one-thousand-three.” He then turned the wheel to the right, gained control of the plane, and leveled off. He had prevented the crash.
Now it was Hall’s turn. When the rudder suddenly went hardover, Hall followed Hewett’s instructions and quickly turned the wheel to the right. “Hold it! Hold it!” Hewett told him. The simulator started to plunge toward the ground, but he stopped the roll and brought the nose back up. “Ease it out,” Hewett said. Hall had saved the plane.