Flight 232: A Story of Disaster and Survival (47 page)

A technician sat at the electron microscope at GE and pinged the inner shells of electrons inside that pit. Only this time he used a more sensitive test than the energy dispersive spectography. With EDS, they had seen the energy of the X-rays dwindle away and not give a clear signal of what was in the pit. The technician switched modes to use X-ray wavelength dispersive spectroscopy (WDS). Cherolis said, “WDS uses a different type of detector to measure the wavelength of the X-rays, and the wavelength tells you the energy.”

The X-rays coming off of that area were put through a device called a diffractometer, “a crystalline substance with a known atomic spacing,” Wildey explained. “Because of the way that the X-ray interacts with a crystal, it will diffract only at specific angles.” In WDS, the diffraction angle then told him, even at low energy levels, what atomic element he was looking at. It is so accurate that it also told him the amount of that element. “EDS is used whenever possible because it is simpler and faster, and for many elements it is perfectly okay. For specific elements, EDS cannot distinguish the energy peak from two different elements when they are close together, and is not really good at the low energy X-rays. In these cases, and for more precise measurements, WDS is used.”

Once the technician had subjected the sample from disk 00385 to the WDS procedure, the wavelength readings came back loud and clear: the pit was full of nitrogen, and not in trace quantities either. The
serviceable metal cast by TIMET in 1971
had contained .008 percent nitrogen, an allowable level. In this defect, the percent by weight of nitrogen went as high as 2.07. Moreover, where the nitrogen was at its highest, the aluminum and vanadium were at their lowest, destroying the properties that make the alloy useful. It was conclusive: the titanium had been contaminated with nitrogen when it was first melted at TIMET in Henderson, Nevada, in 1971. When titanium is combined with enough nitrogen, it becomes a ceramic material. Even at a concentration of 2 percent, nitrogen makes titanium brittle. Those brittle “inclusions,” as they’re called, can fracture and fall out. Then the little hole or pit that’s left can start a crack that begins to grow outward with the centrifugal pull of the spinning fan. Wildey’s team performed another test with an indenter to measure hardness. The material in the pit was far harder than titanium alloy should have been. That’s why the defect was known as hard alpha. Titanium stretches and bends. Hard alpha breaks. The original spot of hard alpha had fallen out, leaving a pit that was about 1.4 millimeters (about .06 inch), not quite the width of a grain of rice. The full name the metallurgists use for such a defect is nitrogen-stabilized hard alpha inclusion.

Charles Martz came out the back of the fuselage, leaping through fire, and set foot on the soft earth beside the runway. In a speeding, almost giddy, daze, rejoicing in his bittersweet survival, he stepped up onto the concrete surface. He could scarcely believe he was alive. He raised his eyes to look out over the sunburned, smoky landscape and saw “the goddamndest mess I’ve ever encountered, like a small battlefield.” He saw at least a dozen people still in their seats. Their clothes were torn or blown or burned from their bodies, “completely naked in front, missing limbs, missing faces, some breathing, some moaning, and others just deader than a door nail.” As he walked along the runway, he came upon a United Airlines pilot. “He tried to sit up,” Martz said. “I saw a huge triangular hole in his forehead and I told him to just lie still and that help was on the way, but it was too late for him.” Martz walked on and encountered a lady whose hair and clothing were burned away, her stockings melted onto her legs. Next a man completely naked in front, black from soot. “Finally, I will never forget stopping near a lovely young girl still strapped to her seat, breathing slightly. Her blouse was white, her slacks were blue. At the end of the trousers were two snow-white ankle bones where her feet used to be. I had never seen the whiteness of bones that are freshly exposed like that. Her beautiful blue eyes were wide open, staring at the sky. I spoke to her. Nothing at all, as the blood drained from her body.” At this point in his description, Martz groaned audibly with the pain of dredging up those memories, and said with regret, “She was just a
pretty
gal.”

Only a minute or two had passed. He turned back toward the burning fuselage and saw that a fire truck was spraying foam on the wreck. He saw two men come out of the flames. Larry Niehus sprayed them with foam.
Two priests dressed in the protective gear
of fire fighters entered the burning fuselage to administer the last rights to some of the dead before the heat and melting structure drove them out. As Martz turned away, aware that he could no longer help those people, a pickup truck pulled up beside him. “And here I am, I’ve got my tie on, I’ve got on a blue oxford button-down shirt, pen and pencil still in the pocket, and looking like I’m ready to sell insurance.”

The driver of the truck looked at him and said sternly, “What are you doing here?”

“Well,” Martz said, jerking his thumb over his shoulder, “I just came in on Flight Two-Thirty-Two.”

The driver thought Martz was an ambulance-chasing lawyer, but Martz convinced him that he really had been on the plane that lay burning beyond the trail of the dead. As Martz boarded the truck for the ride to the triage area, he realized that he was completely uninjured except for that small burn on one finger. He hiked over to Graham Aviation and called his wife Janie. That night, along with Martha Conant, Rod Vetter, John Transue, and others, Martz boarded the United Boeing 727 to Chicago. He experienced no anxiety about flying. “We’re not going to crash again. I was sure of that.” As an experienced fighter pilot, he also took comfort in seeing the old pilot who flew them to Chicago that night. “He was a crusty old bird, they were very smart about that. He had the look of a trusted aviator who was going to get you there safely.”

In midair, he crossed paths with the Grumman Gulfstream carrying the NTSB Go Team. He didn’t reach the Westin Hotel at O’Hare Airport until two in the morning. Once in his room, he tuned the television to CNN. “That was the first time I’d seen the video of the crash,” he said. “And that’s when I started to shake. I had been relatively calm until I saw that video, and I said, ‘Jesus Christ, how could anybody survive that?’ I got up, walked over to the bar in my room, and downed two quick shooters of vodka.” Then he went to the mirror and looked at the face in the glass. “It was mine. It was alive.”

In bed at last, he slept soundly. He rose at six in the morning, caught the shuttle to the airport, and boarded the first flight back to Denver. United had planned a flight for survivors later that morning, but Martz wanted to get out of there. He was going to put the crash behind him and get on with his life. He needed to see Janie.

CHAPTER TWENTY-THREE

W
hen a fatigue crack grows due to vibration
, it may be put under strain millions of times. In the cult of metallurgy, such a phenomenon is known as a high-cycle crack. Fan disk 00385 was put under strain each time it flew, so the total number of times the crack grew was in the thousands. It was therefore known as a low-cycle crack. Wildey described the characteristics of a low-cycle crack: “It’s flatter. It’s smoother. And it’s striated in that it has these clamshell marks or beach marks, as they’ve been referred to”—metallurgists at General Electric used the term
river marks
—“where the crack, as it progresses, will leave behind these growth rings that are the sign that the crack is propagating under tensile stresses over time.” To Wildey’s trained and unaided eye, from five feet away, he couldn’t see individual striations—they were too small—but he could see the overall way the surface reflected light in an undulating pattern. The striations ranged from more than three hundred thousand per inch near the pit to about two thousand at the far edge of the area of fatigue. The crack, in other words, had started by growing 1/300,000th of an inch on each flight.

“That’s hardly anything,” said Cherolis. “But it’s moving.” And it’s accelerating. By the end, it was growing by one half of a thousandths of an inch on each flight. “Freaky, isn’t it?” he remarked.

Cherolis and Pridemore went through the photos taken in the electron microscope and counted the striations, as you would with the growth rings in a tree. “We don’t count every single one of them,” Cherolis said. “Because it would be very hard to keep track of exactly where you were. So we take snapshots at intervals across the fracture and mathematically connect them up.” Cherolis and Pridemore each counted the fine lines on the same photographs and then compared their results to make sure that they didn’t diverge by too much. Of Pridemore, Cherolis said, “He’s a guy who really knows how to do striations.”

They then plotted the density (cracks per inch) versus the depth (distance from the pit). They fitted a curve to the data to produce a beautiful power law diagram and then calculated the area under that curve. “I think we used Mathcad at the time of Sioux City,” Cherolis recalled. (More recently, they employed a computer program called Excel.) “There is also a graphical method you can use that is quite accurate, breaking the area under the curve into a series of rectangles and triangles. I usually do this to make sure I didn’t input anything wrong into the software for integration. You do not just stop at what you need to prove a defect or primary cause,” he added. “In a case this important and significant, you go over and above.” All of these efforts were aimed not merely at learning what had gone wrong but at preventing accidents in the future.

Cherolis knew well what he was looking at in this hard alpha defect and the crack that grew from it. “The crack accelerates its growth as the crack grows deeper into the metal”—that is, the farther it grows out from the center of the disk, the faster it grows with each flight or cycle. When they had calculated the area under the power law curve, it came out to about fifteen thousand cycles. Once Cherolis and Pridemore had that number, they asked how many times the disk had flown since 1972.

By that time, Young, Moehring, Cookson, and others from the NTSB and GE and United had researched the records deeply, and they all knew: fan disk 00385 had flown 15,503 times. At the time of its last fluorescent penetrant inspection, that disk had flown 14,743 times since its debut in early 1972.

“I was pleased at how close it was to the real number of flights,” Cherolis said of his and Pridemore’s calculations.

When I asked Cherolis if someone should have seen the crack in 00385, he said, “Ah. That’s the tragic thing. Somebody’s probably losing sleep over that.” If that crack was on a test bar, “no one could miss it,” Cherolis said. “It’s a piece of metal. It’s asked to do a certain job. And you do everything you can to make sure that it’s as good as it can be. You just hope the airplane’s on the ground or just taking off and they can reject the takeoff.” He added, “It’s sad. But that’s just how things work. And it still is that way. What’s less likely is that a hard alpha is going to be in titanium with the newer processes” for making that metal.

The materials used in making fan disks are much better today than they were in the 1970s. The process of making titanium has changed radically. For example, while the billet used to make fan disk 00385 was sixteen inches in diameter, billets in use by the time of the crash were only ten inches, making it easier to detect flaws. In addition, GE changed its specification to call for makers of titanium to use such exotic techniques as electron beam cold hearth melting and plasma torches. As a result, hard alpha defects are less frequent. In fact,
by 1990, GE claimed that the electron beam
“melting process addresses the residence time limitation of VAR [vacuum arc remelting], permitting assurance of dissolution of any hard alpha inclusions which may be present in the raw material.”

I asked Wildey, “Could this accident happen today?”

“Yes,” he said.

Dave Randa and I had finished our lunch at Fox Fire, the bar and grill in Geneva, Illinois. He was telling me his impressions from the time immediately after the crash. Since he and his mother Susan had been splashed with blood, a nurse at the hospital called a friend of hers who also had a nine-year-old son. She brought a pair of yellow shorts and a blue T-shirt to the hospital for Dave to wear. The nurse gave Susan a pair of her surgical scrubs. Then Dave and Susan were taken to the dormitory at Briar Cliff College. “But the biggest thing was that I had lost my Cubs hat,” Dave said.

The dorm was overrun with reporters, and Dave commented to one of them that he had been on his way to a Cubs game and had lost his favorite Cubs hat. When Jim Randa arrived in Sioux City, he drove his wife and son to the house of Susan’s brother in Chicago. The Cubs heard about the lost hat and sent over a package with a new hat, as well as signed photos of Dave’s favorite players, Andre Dawson and Mark Grace. They also sent ten box-seat tickets, “second row right behind the batter’s box,” Dave said. “Unbelievable tickets. Absolutely loved it. Beautiful day. You could reach out and almost touch the players. When you’re nine, that’s amazing. And at that time, it was so hard to get that type of ticket. Vance Law, I want to say, hit two home runs. Got to meet him. Great person.” Dave remembered correctly. Law hit one homer off of Jeff Brantley and the other off of Atlee Hammaker. “They really took care of us. The locker room after—Andre Dawson and Mark Grace were so nice. They were just spectacular.” A brilliant Saturday afternoon in the friendly confines of Wrigley Field, and on that glorious day the Cubs beat the San Francisco Giants 5 to 2. Life did not get any better than that, Dave thought.

By the time the Randa family returned to Colorado, however, Dave was having nightmares and flashbacks and couldn’t sleep for long. “I’d wake up scared of the plane crash, scared of death, scared of somebody breaking in.” He was now afraid of things that had never entered his mind before.

Susan said, “Dave was afraid of everything. He slept on the floor of our room for like six months.”