Mayday Over Wichita (17 page)

Interval

Interval

Later that evening, when Szmuc's crew landed at McConnell, they requested the Boeing mechanics take a look at the autopilot and repair it. Nevertheless, Boeing's contract with the air force allowed it to fix
only
minor maintenance problems, and since the autopilot was more complex, Boeing was not required to do so. Therefore, when Szmuc and his crew departed McConnell on that fateful Saturday morning, the autopilot was still malfunctioning.
³³¹

An electronics equipment expert for the air force, Sam E. Taylor, testified before the investigation board that the autopilot malfunctions on Raggy 42 were “abnormal” and “way out of reason.”
³³²
Although the air force had not grounded a plane for such cause before, it was clearly a problem and point of concern for the crews based on their complaints.
³³³

Boeing, as expected, was less forgiving in its analysis of the pilots, denying any malfunctions on the KC-135 it built. Its general counsel, George Powers, stated, “…[T]hey went up at a low altitude and immediately started a left turn too sharply over the university. They started side slipping…and didn't have enough speed to make the turn. His left wing was too low and he could not pull it up. That's what happened and there was never any question about it. Pure pilot error.”
³³⁴
Powers's statement would be refuted in time.

14

A SUBTLE KILLER

Decades after Raggy 42 came smashing down on Piatt Street, three nearly identical crash investigations were headed up by the National Transportation Safety Board (NTSB) in the 1990s. These investigations provided a new perspective on why Raggy 42 and other similar aircraft disasters occurred in the manner in which they did and what was to blame. The search for answers included all the trappings of a police investigation, with ardent detectives methodically chasing an elusive serial killer, who strikes without warning and then disappears into the shadows, leaving no trace. It took the NTSB ten years of fact-finding to locate the cause behind these three occurrences, and since that time, numerous lives have been saved because of the information it identified. There was one striking similarity in the explanations that investigators finally gave as the root cause for these crashes: unscheduled rudder deflection.

T
HREE
F
LIGHTS
, T
HREE
P
ROBLEMS

United Airlines Flight 585

The longest crash investigation in aviation history started with United Airlines Flight 585, a passenger flight carrying twenty-five souls from Denver to Colorado Springs, Colorado, on March 3, 1991. Four miles south, at 9:44 a.m., Flight 585 was coming in to land on runway thirty-five at Colorado Springs Municipal Airport. The plane was piloted by Capt. Harold Green and First Officer Patricia Eidson, both seasoned pilots, who had just been cleared by Air Traffic Control to make their final approach when disaster struck. According to the NTSB Aircraft Accident Report, a blood-curdling scream of “Oh no!” was heard on the flight voice recorder just before impact.
³³⁵
Those were their last words. Investigators determined through eyewitness accounts that, once the plane had completed its final turn and headed in for its approach, it “rolled steadily to the right and pitched nose down until it reached a nearly vertical altitude before hitting the ground.”
³³⁶
The crash was extremely abrupt. Within ten seconds, the pilots experienced some minor turbulence, and the plane mysteriously spun out of control, turned upside down and spiked nose-first into the pavement, leaving almost nothing of the 737 to investigate.

An impact crater, much like the one found on Piatt Street, was virtually the only evidence. And because the plane did not skid upon impact, the fire was contained in a condensed area—also like Piatt Street. The five crew members and twenty passengers onboard, unfortunately, too, were immediately consumed by the inferno of jet fuel ignited by the crash.

Determining the cause of Flight 585's sudden catastrophe seemed nearly impossible. The defense of pilot error, it was later found, did not hold up. Both the pilot and first officer were experienced professionals, and investigators found no indications that they would have erred to this magnitude. Mechanical failure was also a difficult theory to substantiate, since the 737 had an impeccable safety track record, and indicator needles on the hydraulic pressure gage showed the plane's engines were running normally upon collision. Turbulent winds were present, but that theory soon dissipated as investigators determined that “too little was known about the characteristics of rotors [a horizontal axis vortex] to conclude whether a rotor was a factor in the accident.”
³³⁷

The flight controls were also examined. And although small metal chips were found inside the hydraulic fluid of the plane's power control unit (PCU)—a dual servo valve that disperses hydraulic fluid to move the rudder—investigators were still uncertain as to how this could have crippled the plane. Fresh out of clues, and unable to determine the cause of the accident, the NTSB was forced to give its final assessment on December 8, 1992. It regrettably stated that investigators “could not identify conclusive evidence to explain the loss of United Airlines Flight 585.”
³³⁸

US Air Flight 427

Three years later, on September 8, 1994, at 7:00 p.m., US Air Flight 427, also on a Boeing 737, was approaching Pittsburgh, Pennsylvania, carrying 132 souls on board. The plane, piloted by Capt. Peter Germano and First Officer Chuck Emmett, had made its final approach into Pittsburgh International Airport. All seemed calm when, suddenly, the jet rolled sharply to the left as if it had a mind of its own. The cockpit voice recorder transcripts captured the frightening last words:

Then silence. All 132 lives were lost.

The plane slammed into a hill six miles northwest of the Pittsburgh International Airport near Aliquippa, Pennsylvania, leaving fractured pieces of debris and human carnage in its wake. The investigators were forced to don biohazard suits due to the shroud of human remains scattered along the hillside.
³⁴⁰

Finding the cause of this crash was yet another mystery: very little was left of plane; both the pilot and first officer were skilled and had excellent track records; the 737's safety record gave no indication that mechanical failure existed; and the engines had been functioning properly prior to impact. The one thing the investigators could examine was the plane's tail section. Again, the investigators found small metal chips in the hydraulic fluid of the plane's PCU servo valve, leading them to believe that a problem had arisen whose cause was still eluding them. The PCU was disassembled and subjected to numerous tests. Still, they came up empty-handed. They were no closer to finding the elusive killer. And just like after the Piatt Street crash, public dissatisfaction grew as the investigation waned.

Eastwind Airlines Flight 517

Two years later, on June 9, 1996, Capt. Brian Bishop and his first officer were making their final approach to Richmond, Virginia, in a Boeing 737 with fifty-three souls on board. In the official interviews taken after the salvaged flight, Capt. Bishop explained how, as they descended to approximately four thousand feet, “the airplane yawed abruptly to the right and then rolled to the right.”
³⁴¹
Attempting to correct the yaw to the right, the captain instantly stomped on the “opposite rudder and stood pretty hard on the pedal.”
³⁴²
According to the first officer, the captain was “fighting, trying to regain control” while simultaneously “standing on the left rudder.”
³⁴³
He also cut the yoke to the left and increased the right throttle to try to pull out of the deadlock. The plane, however, refused to budge and remained on its side for half a minute before releasing. Then it happened again.
³⁴⁴
Capt. Bishop told his first officer to “[d]eclare an emergency” with Richmond Tower and “[t]ell them we've got a flight control problem.”
³⁴⁵

Staring death in the face, the pilots did what they were trained to do: they started performing their emergency checklist—something the other pilots, with only seconds to spare, had had no time to do. Then, as quickly as it began, the rudder finally released its grip. The plane did not roll onto its side for a third time, and the pilots were able to land the plane safely at Richmond. The NTSB, Boeing and US Air investigators immediately secured the plane and the crew so they could determine why Flight 517 defiantly rolled over.

The investigation combined Flight 517's rudder issues, pilot testimony and what little knowledge they had gained from the prior accidents involving Flight 585 and Flight 427. At first, it seemed as though the investigation would again end in a standstill. But when the investigators began thermal shock testing with the PCU on Flight 427, they finally had a breakthrough.

Thermal shock testing uses “hot hydraulic fluid injected directly into a cold PCU to explore the effects of extreme temperature differentials on the main rudder PCU's operation.”
³⁴⁶
They discovered that, if they soaked the PCU in dry ice and then injected it with hot hydraulic fluid, it jammed. When the valve jammed, it reversed. This meant that if the pilot tried to correct the yaw by pressing down on the rudder pedals, the rudder moved in the opposite direction autonomously. The mystery was solved. When the PCUs malfunctioned due to thermal shock, they jammed the rudders, causing Flight 585, Flight 427 and almost Flight 517 to crash.

The official investigation report from Flight 585, later amended, found that all three planes had experienced malfunctions due to unscheduled rudder deflection and had little time to recover from it.
³⁴⁷
The NTSB determined the probable cause of the US Air Flight 427 accident as “a loss of control of the airplane resulting from the movement of the rudder surface to its blowdown limit.”
³⁴⁸
The report concluded: “The rudder surface most likely deflected in a direction opposite to that commanded by the pilots as a result of a jam of the main rudder power control unit servo valve.”
³⁴⁹
In light of the culprit, Boeing made changes to the valves following the tests. Subsequently, there have been no recurrences of that nature.

This is not to suggest that Raggy 42 experienced thermal shock that jammed the rudder. The KC-135 is a much different plane than the Boeing 737 involved in these three cases. Furthermore, these accidents occurred during landing and not during takeoff, when Raggy 42 malfunctioned. Even so, the three flights investigated by the NTSB provide another viewpoint on how a plane and its pilots respond to an unscheduled rudder deflection.
³⁵⁰
If highly trained pilots in the 1990s, nearly three decades after the Piatt Street crash, had difficulties with unscheduled rudder deflection, it is plausible that Captains Szmuc and Widseth, flying a relatively new jet plane in 1965, would have experienced similar complications for which they were not prepared.

“Pilots would be more likely to recover successfully from an uncommanded rudder reversal,” said the Flight 585 accident report, “if they were provided the necessary knowledge, procedures, and training to counter such an event.”
³⁵¹
Chester I. Lewis offered his compendious judgment after reading the accident investigation report on Raggy 42: “Boeing neither in its flight handbook nor otherwise informed the [air force]…of the flight characteristics or procedures to be followed in actual flight if a power rudder has a severe deflection as it did in the case herein.”
³⁵²
The findings of the Air Force Accident/Incident Investigation Report verified the slim chance the pilots had of recovering Raggy 42, saying, “To sustain flight with maximum rudder deflection at speeds approximating Raggy 42's would have required asymmetric power adjustment, flaps 20 degrees, and exceptional pilot technique.”
³⁵³