Flight 232: A Story of Disaster and Survival (46 page)

Floyd Brate was an expert in a technique called replication. Everyone working on the analysis wanted to extract as much information as possible from the disk before they began the process of destroying that tiny pit, which would be the last step in the analysis. Brate would make exact replicas of the fracture out of several different materials. He first washed the part with deionized water. Then he wet acetate tape with acetone to soften it, and covered the fracture with it so that it would adhere to the surface and mold to all the fine features of the metal. When it had solidified, he pulled the acetate tape off and repeated the process several more times.

“I did pull acetate tapes off,” Brate said, “and we lifted fluorescent penetrant off of that, which was analyzed on a scanning electron microscope. So there was a prior crack there where FPI penetrant seeped in the crack.”
General Electric manufactured disk 00385
from September through December of 1971, and then installed it on an engine, which was sent out to Douglas in Long Beach and hung on a DC-10. Fan disk 00385 was removed from that engine and inspected in September of 1972.
It was then installed on various engines over time
, and it was removed and inspected again and again: in November of 1973, in January of 1976, in June of 1978, in February of 1982, and in February of 1988. At each of those times, the disk was subjected to fluorescent penetrant inspection. The process is simple and seemingly foolproof in theory. (Practice is another matter.) The inspector submerges the disk in—or sprays the disk with—a yellow-green oily liquid. “It has to sit for half an hour or more,” said Cherolis. If there are any cracks in the metal, the dye will seep into them through capillary action. Then the inspector washes off that penetrant fluid, being careful not to wash it so aggressively that he washes it out of the cracks. He then dusts the disk with a powder, called a developer, that acts as a blotter and draws the fluid out of the cracks to make them more readily visible. Under ultraviolet light, any crack would fluoresce yellow-green. The FPI inspection in 1973 was done at the GE Airline Service Department in Ontario, California. The other five inspections were conducted at the United Airlines CF6 Overhaul Shop in San Francisco. In early 1988, someone noticed that the engine carrying fan disk 00385 showed some corrosion in another unrelated part. The engine was disassembled, and while the fan disk was off of the engine, it was subjected to its last fluorescent penetrant inspection.

The inspector coated the disk with the dye, and the dye should have shown the defect, as well as the fatigue crack that had grown from it over the years. By that time, the crack had grown to about half an inch long and a quarter inch deep. By aerospace engineering standards, it was large. It should have attracted the attention of anyone looking at it.

Cherolis summed it up: “During that last fluorescent penetrant inspection of the fan disk at the airline shop, if everything was done right, a bright yellow-green line one half inch long would have appeared on the part under black light. So it would be like a glowing line drawn in the part well within the detectable range. Everyone who has tried to perform that inspection on a real part that is over 300 pounds can see how it might be missed. So the event was a combination of the defect, the lack of detection over many inspections, and the design of that airplane all coming together.” However, the uninitiated person, imagining a bright fluorescent line on a piece of metal, cannot so easily see how it might be missed. One would think that it should have been obvious. Yet it was not.

“And I don’t know whose fault that was,” Brate said. “We lifted the penetrant off on the initial first two replicas.” A process called secondary ion mass spectroscopy was used on the acetate tape impressions that he had made. The test detected molecules of the various phosphorous compounds used in the fluorescent dye. Gas chromatograph mass spectroscopy was also used on the wash water from the ultrasonic cleaner. Those phosphorous compounds were detected in that water as well. In the years since the crash, there has been some debate about where those phosphorous compounds came from and whether the NTSB detected dye in the crack or not. In sworn testimony during the NTSB hearings, witnesses said that no phosphorus was found and no shot peening had occurred inside the pit.

In 2013, Robert MacIntosh said, “
The Public Hearing testimony
was conducted as a fact-gathering exercise and before all the factual evidence was able to be assembled. Later, in the end process of writing the final report, as I recall, Jim Wildey had evidence based on reliable observation on both the FPI fluid and the multiple cracking within the shape of the pit.”

Wildey responded, “I personally was never convinced that there was any evidence one way or another about the presence of phosphorous compounds (from the inspection process) on the fatigue region. However, the fatigue region itself did contain surface features that corresponded to the size of the crack at the last inspection. These features were duplicated in tests that GE performed and clearly show that the crack was open during the last United inspection and capable of sucking up the FPI fluid. One of the main arguments presented by United was that the bore surface had residual compression stresses that closed the crack and prevented penetration of the FPI fluid. This is clearly not true, based on the appearance of the fatigue crack region.” Wildey’s team could mathematically calculate how big the crack was at the last inspection by knowing how fast it was growing over time. And this calculation told them that the crack was large enough to take up the penetrant dye.

This sort of detailed technical argument is of purely academic interest to engineers and metallurgists, since no one disagrees that the crack was there and was not detected. In the end, no one was willing to blame the inspector for missing it. His boss told me that the man had suffered terrible emotional symptoms. In fact, a
concerted effort was put into keeping the technician’s identity secret
. It has never been released. His identity was concealed by a number. Each inspector had a stamp with a unique number used in signing off his work. David Cookson, in charge of researching the records for United Airlines, researched that stamp number and discovered not only the inspector’s name but also the fact that he knew the man personally. “I researched the number. I’m sorry I did. To me at the time—just a personal note—it was somewhat upsetting to find out I knew the person who had accomplished the inspection. . . . I can imagine he would have problems. And that’s what troubled me. The guy was doing his job. And he did it to the best of his ability. There were defects in the disk prior to his inspection. But I’ll never say that he missed it. I never will say that. Because I don’t know. I don’t know what the circumstances were. . . . Unfortunately, he happened to be the guy that inspected the disk.”

After Floyd Brate had made replicas of the crack and had detected the fluorescent dye, technicians used a water-cooled saw with an abrasive blade to cut off the bore end of the long piece they had removed from the fan disk, including the suspect material in and around the pit. The team now had a piece small enough to fit into the scanning electron microscope and could begin looking at the crystalline structure of the metal in fine detail.
In preparation for this step, according to Cherolis, a technician named Ivan Miller put the part
, which now looked like a piece of abstract jewelry, into an ultrasonic cleaner to remove any oil or dirt that still might be adhering to the fine structure of the metal. Then he put the piece into the electron microscope and turned on the vacuum pumps to evacuate all the air so that the electron beam didn’t burn up the filament. In the darkened room, the screen was soon glowing a winter bluish white with the image, as he dove down and down, as if through living tissue, into the structure of the titanium. The day was long, but in the end, Cherolis, and Pridemore were looking at finished photographs from the scanning electron microscope. They also used the scanning electron microscope to perform electron diffusion spectroscopy, to look down into the electron shells and see what sort of atoms were in that pit. But as had happened with the sister disk 00388, the signals from the foreign material were too weak to read.

“Unfortunately,” said Wildey, “nitrogen is one of those elements that’s very difficult to detect in that process, so we were still a little mystified or a little uncertain as to what was going on there.”

As United Flight 232 approached Sioux City and the crew began telling everyone to brace for the worst, Charles Martz sat looking out the window and said to himself, “Well, this is the only crash I’m ever going to be in after about three thousand hours of flying. So I’m going to sit here and look out the window and see it.” Martz sat four rows behind Jerry Schemmel, three rows ahead of Schemmel’s best friend Jay Ramsdell. Martz was sitting on the boundary that would separate the two friends forever. He was furious, and rightly so. He turned to Luella Neubacher and told her to go ahead and brace, and he’d watch out for her. As the plane descended lower and lower, “I could see the airport off the right wing as we sort of wobbled to the right. We were descending in an erratic fashion. It was easy to see that we were going way too fast and descending too rapidly.” In a normal landing, the plane comes down at three hundred feet a minute. This DC-10 was approaching the ground at more than sixteen hundred feet a minute.

Resting his head on the pillow and looking out the window, he could hear “the desperate pounding of my heart in my ears as the adrenaline took over.” The green cornfield rushed by, “almost a blur.” At that point, Martz heard both engines rev up. “But the right wing tip hit the dirt, and the landing gear departed.” He recalled that when he saw that wing tip hit the ground, “finally, I closed my eyes and mentally kissed my ass good-bye.” At that point, he thought, “It won’t be long now. Something’s going to hit me in the head and it will all go black.” He said, “I remember being whipped back and forth and banging against the cabin wall on the right, careening down the runway, flailing about, waiting for the final blackout, the last fragile moments of life, wondering, waiting, furious, frightened, and alone. And finally everything did go black.”

When the noise and motion stopped, he enjoyed no moment of reprieve, no time to think, as some passengers had in B-Zone. Choking smoke began to fill the aft cabin immediately. “There was an instant of silence and then a surge of moans and screams as the injured and terrified realized what had happened,” Martz said. “I was awake the entire time and was truly amazed to find that I was still alive and able to think, move, and breathe.” Within seconds, however, the smoke in the aft cabin grew, “deep black, acrid, utterly frightening,” he said. “You had to hold your breath for as long as you could.”

Martz had an advantage over everyone else in C-Zone. As a Navy fighter pilot, he had been trained to get out of a cockpit while inverted, in the dark, on fire, and even under water. When the screeching of metal stopped and he heard the people around him begin to scream and moan, he looked at Luella Neubacher hanging from her seat belt beside him and decided that he’d release himself, then release her, and they’d get out together. He released himself but retained no recollection of doing so. His Navy training had taken over. He tried to look up to locate his seatmate, but he could see “nothing except the floor lighting running left and right immediately above my head.” He saw no people. “I looked to the right and everything was darkness. I looked to the left at about eye level and saw flames outside through a window. I stood up and kicked at the window. There’s no way to kick a window out of an airplane. I looked further to the left and saw a fire blazing at the open end of the fuselage, perhaps five feet away.” He decided that he would rather suffer severe burns than die of smoke inhalation.

More than two decades later, Martz said, “I did something that I’ve always regretted, and I carry that with me to this day. It was just a sad, sad thing, because I fumbled around trying to find Mrs. Neubacher, and I couldn’t find anybody. I said, well, there’s fire to the rear. I figured where there’s fire, there’s oxygen.” He moved toward the opening and ran into a man who stopped him and said, “It’s burning out there.”

“I know,” Martz said. “Let’s go.”

“No way,” the man said.

Martz pushed past him. He never saw the man again. Martz now ran toward the flames, gaining as much momentum as he could on the uneven surface beneath his feet. He leapt into the flames. “It was wide open,” he said with a note of astonishment in his voice. He had literally jumped through fire to freedom, while those around him perished. He burned a finger going through but was otherwise untouched.

“Suddenly I was standing in a muddy cornfield outside the airplane in bright sunshine.” He walked a few steps and stopped, thinking, “It’s not that bad. Maybe I can go back in and get that lady.” But, he continued, “At that moment there was an explosion and a burst of fire to the right . . . as more fuel fed the fire.” He thought, “Charlie, you’re alive and well, there’s no use in being a dead hero going back into that inferno. Get the hell out of here.”

The oxygen bottles and fire extinguishers began exploding. Martz said he was thinking, “I have survived the crash. The cabin is filled with smoke. Stuff is exploding.” He walked farther away from the wreck, torn by indecision.

“And I finally said, Forget it.” Of course, he could not forget it, not even when he spoke to me at the age of eighty-one. With her favorite gray coat on, ready to walk out with him, Luella Neubacher died of the smoke.

James Wildey, working at the GE lab, cut a half-inch piece of the defect and mounted it in Bakelite plastic to make it easier to handle. He then held the specimen by hand, pressing it against a rotating circular platform with finer and finer grits of diamond or alumina, “until you get down to stuff that is so fine I can’t even describe it to you, to make a mirror surface on this metal.” It took about an hour of polishing to achieve that finish. This was the final piece of physical evidence, and he was destroying it little by little to demonstrate the ultimate cause of the crash of 1819 Uniform. The process was called step polishing, and it involved gradually grinding away the defect in the metal and then polishing it and testing it at each step using the electron microscope to determine what atomic elements it contained.