Air Disasters: Dramatic black box flight recordings

AIR DISASTERS

DRAMATIC BLACK BOX FLIGHT RECORDINGS

MALCOLM MACPHERSON

Collins

Table of Contents

Cover Page

Title Page

Introduction

Colorado Springs, Colorado, USA 3 March1991

Memphis, Tennessee, USA 7 April 1994

Charlotte, North Carolina, USA 2 July 1994

Pittsburgh, Pennsylvania, USA 8 September 1994

Eight Miles Off East Moriches, New York, USA 17 July 1996

Monroe, Michigan, USA 9 January 1997

Nantucket, Massachusetts, USA 31 October 1999

Marsa El-Brega, Libya 13 January 2000

Off Anacapa Island, California, USA 31 January 2000

Rancho Cordova, California, USA 16 February 2000

Linneus, Maine, USA 19 July 2000

Charles De Gaulle Airport, Paris, France 25 July 2000

Aspen, Colorado, USA 29 March 2001

Near Burdakova, Russia 4 July 2001

Belle Harbor, New York, USA 12 November 2001

Taos, New Mexico, USA 8 November 2002

Charlotte, North Carolina, USA 8 January 2003

Jefferson City, Missouri, USA 14 October 2004

Chicago, Illinois, USA 8 December 2005

Philadelphia, Pennsylvania, USA 7 February 2006

Lexington, Kentucky, USA 27 August 2006

Index

Copyright

About the Publisher

INTRODUCTION

Before the publication in 1984 of my book The Black Box I had called at the offices of the US National Transportation Safety Board (NTSB) in Washington, DC, on the unlikely chance that the Board made available to the public transcripts of commercial air transportation cockpit voice recordings (CVRs), which I knew about from commercial pilot friends who read them out of personal interest and had handed them on to me. My request that day surprised the Public Affairs officer who said that no member of the public had asked to see them before. He wondered out loud what anyone would want with them, as he led me over to thirty or forty transcripts that someone in the office had patiently typed out, copied and stacked up against filing cabinets, looking as if they were ready to be thrown out.

I told him that they fascinated me in a way I could not describe even to myself but their hold on me lasted long after I read them. Maybe they would interest other like-minded people if they were made available in book form. Come to think of it, he told me, he read the transcripts, too, and not for strictly professional reasons. He could not articulate his interest either, but we sat on the office floor and, from memory, he chose twenty of the transcripts that he thought would be of most interest to me. Before long I staggered out of the building under the weight of at least a score of the transcripts. As soon as possible I set about turning the pages into a form that would make sense to readers like me with no specialized knowledge of flying or of the mechanics of commercial aircraft and aviation. The Black Box, as a specialized book category, was born.

Now, nearly thirty years on, the aviation industry has changed remarkably, as anyone who flies knows. I must say that most of these changes are for the worst from the passengers’ perspective. Today, already cramped spaces in aeroplanes are smaller, meals are worse or in some cases nonexistent, delays are longer, cancelled flights are stacked up higher, rude and indifferent behaviour by officials seems almost routine, and more people like you and me want only a quick and merciful end to the experience of flying between home and destination that does not involve a crash. In 2007 passengers’ complaints to the US Department of Transportation increased by 60 per cent, which means that, for instance, a typical passenger from Washington Dulles Airport was angry enough with the service to complain once every day and a half. In an editorial in the spring of 2008 the Washington Post had this to say: ‘Air travel has gotten so bad these days that going to the airport requires either an exercise in sadomasochism or an abiding faith that everything will be okay. That faith seems to be shattered daily.’ Beneath the surface passengers are seething; to tell the truth, flying is no longer the convenient way to travel, but in America it is the only way, at least with gasoline prices inching up, train services starved of federal funds and bus travel nearly nonexistent. Bad as it is, we have no other choice but to fly.

That said, in one very important—the most important—respect, changes over the years in commercial air travel have been excellent for passengers: whenever we set out to reach a destination by air, we arrive. In other words, though flying might be miserable it is safe, pure and simple, and, remarkably, it is getting more so all the time. As if air safety were a warp zone of science fiction, safety managers for airlines and the government are today reaching beyond known weak links into the realm of ‘what might conceivably happen’ and are making changes for safety before trouble happens, so that passengers never have to experience incidents that did not have a chance to develop.

Indeed, there has been no major aircraft crash in the last seven years in the USA. That’s safety to bank on. Indeed, 2007 was a typically excellent year. With 4.65 billion air passengers travelling worldwide (769 million on aeroplanes based in the US), 965 people died of all causes in 136 incidents, which were 28 fewer than in 2006, with a 25 per cent decrease in fatalities (and none of these, in fact, included commercial aviation). In the United States 95 per cent of transportation fatalities occur on roads and highways. Waterways and railroad lines account for more fatalities than aviation, which weighs in at a mere 1 per cent of the total for all transportation—or roughly eighty times safer than travel on roads. The time can almost be foreseen when compilations similar to the one you are now reading won’t exist for lack of enough CVR transcripts to put between covers.

Not surprisingly, the safety trends in the United Kingdom, according to the statistics offered by the Civil Aviation Authority (UK), have run closely in parallel with those in the US. While the number of passengers flying to, from or between UK airports more than quadrupled between 1980 and 2006, from 50 million to 210 million, safety has steadily improved to the point of near statistical perfection. In 2006, 185 million passengers arrived or departed the UK on international flights, while 25 million passengers travelled on domestic flights. And yet fatality rates were less than 1 billion passenger kilometres in all years since 1981 and less than 0.1 per billion since 1990. Within the UK, there was one fatality among airline (and air taxi) passengers reported since 2001 and five among airline crews; UK airlines flying outside the British Isles reported no passenger or crew fatalities between 2000 and 2006. (The reason you will find no UK CVR transcripts in this collection is because there have recently been no fatal accidents to record and transcribe, even if the Air Accidents Investigation Branch, part of the Department of Transport, published CVR transcripts or otherwise made them available to the general public.)

The airline industry and government bureaucracies continue to work hard to keep flying safe, aware that fallible human beings will always be involved to create risks. Starting on this new book, I was curious to find out why airline fatalities started to decline, almost precipitously, around 1963. I found that by then the aviation industry had switched over from internal combustion engines to much simpler, and thus safer, jet engines. It was also then that the industry started to build redundancies into aeroplanes beyond just multiple engines. Redundancy, at its simplest level, catches failures. On a two-engine airliner, for instance, if one engine fails on takeoff, the aircraft design calls for the remaining engine to propel the entire weight of the aircraft by itself. Duplication was installed wherever possible.

To this was soon added another safety innovation—the nearly obsessive maintenance of records with which to cross-reference minute and large mechanical and other aircraft failures to forestall other identical (or similar) failures. For instance, when United Airlines DC-10 Flight 232 crashed while attempting to land in Sioux City, Iowa, on 19 July 1989, the NTSB determined post haste that engine failure, and specifically the failure of a single disc in one of the aircraft’s three fanjet engines, had caused the crash. Discs in six engines had been forged from the same alloy ingot, and in a matter of days the other five aircraft with engine discs made from that ingot were identified and grounded, thus eliminating the possibility of a second or third engine failure. That kind of thoroughness is unparalleled.

Over the years starting around 1963, pilot training improved as well. All cockpit crews today train in sophisticated, aircraft-specific simulators. Before simulator use, pilots necessarily acquired experience in the aircraft in real time, which clearly prevented exploration of the aircraft at the edge of its performance envelopes and beyond to see what would happen. Now, with simulators, pilots can crash aeroplanes over and over again while sitting inside a building. Not long ago, I flew ‘aboard’ a simulator of a Chinook military helicopter (MH-47) through an emergency crash/dive that had nearly killed everyone on board in Afghanistan a couple of years earlier. The helicopter pilot in Afghanistan was my pilot in the simulator at a US Army base, and such was the reality of the experience that even the simulator’s technical minders paled; with lights and horns blaring and the cockpit shaking and controls trembling, and, with mountainous Afghan terrain out of the front window, I did not feel that anything was being simulated. I walked away with the reassurance that cockpit crews are the best prepared and trained for any emergency ever.

For decades cockpits were considered authoritarian domains. Like that of a captain on board a ship, the word of the captain in the cockpit was law. Crashes and other incidents occurred when the captain failed to listen to a warning from the co-pilot or first officer. Sometimes, first officers did not even bother to speak up when they saw trouble ahead. Slowly, from around the mid-1980s, cockpit crews began to move away from stratification and develop a new structure in which the cockpit ‘team’ became the ruling entity. First officers were encouraged to speak up and captains were instructed to pay attention to them, for the sake of everyone’s safety. Older pilots who felt uncomfortable in this more democratic setting were asked to resign. The atmosphere improved immeasurably and in terms of safety everyone benefited, including the captain.

At the same time, a vast new array of technology emerged to assist cockpit crews. Technologies that monitor systems, navigate and effectively fly the aeroplane became commonplace in cockpits. And while these tools helped cockpit crews, the individual in the cockpit did not succumb to a natural inclination to rely on technology. Crews today are trained never to let the aeroplane be flown by computers against the pilot’s common sense and experience.

The weak link in safety is man, but that link can also be a last hope.

Taking safety an extra step into a realm where any unknown threat can be isolated and examined before an accident occurs, airlines today are making full use of a technology called Flight Operations Quality Assurance (FOQA), first developed in Europe (and known there as FDM or flight data management). FOQA uses flight data stored in quick-access recorders; the data are analyzed every dozen or so flights. Essentially, FOQA enables airlines to search the data for any events or trends that might signal a conflict with normal, or standard, operating procedures. It affords airlines a real-time audit of what’s going on. This can lead to changes that make flying safer. For example, United Airlines routinely examined FOQA data on its aircraft flying to Mexico City, where it noticed that oddly fast approaches to landing had the potential, at least, of causing runway overrun accidents. United wanted to know what was causing the faster approaches, and FOQA told them. Their aircraft were typically being told to turn early, before a designated intersection. That early turn put the aircraft higher—and thus faster—on the approach. Corrections were made and a potential problem was corrected before it actually arose.

FOQA has also helped to eliminate costly maintenance problems. In one recent example pilots were reporting that they did not know why they were exceeding maximum speeds for deploying wing flaps. The airline analyzed the FOQA data on their aircraft. They reported to these pilots that their speeds were on average only 1 knot or less over maximum. If the airline had gone on the word of the pilots it would have had no other choice but to take the aircraft out of service and disassemble the wing. With FOQA, no such response was necessary and costs were avoided and money saved. Other savings were found through FOQA in gasoline conservation. Conserving only one gallon of AVGAS over a flight by reaching higher altitudes sooner can save an airline tens of millions of dollars a year.

With the recent rapid growth of demand for airline services, new safety concerns will always occur in spite of efforts to anticipate them. One such current concern is called ‘runway incursion’.

While concentrating on making air travel safer in the air than on the ground, airports today are experiencing serious congestion problems on runways and taxiways—specifically, incursions. ‘There are more planes but not more cement’, one person associated with analyzing the issue told me. Congestion leads to incursions, which can lead to crashes and fatalities on the ground. The rate of new incursions has alarmed the US Federal Aviation Administration (FAA) and other nations’ oversight groups. In response, considerable effort is being made to warn cockpit crews directly of impending incursions. At the present time, warnings routinely pass through ground controllers who analyze the specific warnings before alerting the cockpit crews. Those precious lost minutes can make the difference between life and death. The FAA is also studying the benefits of obvious warning lights on runways and taxiways, similar to the lights at road intersections.

Commuter and regional airlines have come under the FAA’s microscope with increased demand for short-hop flights, which feed into major airline hubs. These flights are crashing at rates that worry air safety managers, who have focused their attention on operational crew training of regional airlines. And as more crews receive increased simulator time and intensified qualifying tests and checks, the safety record is bound to improve.

And, finally, to put your minds at rest, there are these astonishing statistics:

Odds of being on an airline flight which results in at least one fatality Odds of being killed on a single airline flight Top 25 airlines with the best records 1 in 6.06 million Top 25 airlines with the best records 1 in 10.46 million Bottom 25 with the worst records 1 in 546,011 Bottom 25 with the worst records 1 in 723,819

But enough about air safety. No reader of this book will have got this far wanting to know more about flight safety. Let’s admit the obvious. This book is unabashedly about air unsafety. The mention of air safety in an unsafety book is a fig leaf for the real reason for publishing CVR transcripts. Over the years I have been editing books such as this some readers have told me that the CVR transcripts actually help calm their nerves, though I do not understand how, since most of the incidents recorded here end with dead bodies and charred aircraft. Truly, these transcripts should give readers the heebie-jeebies. Maybe the claims that nerves are calmed stem from rehearsing a disaster at a distance, imagining what we might do or not do in these same dire straits; the CVR transcripts give some readers the illusion of being in control, when, as passengers, we have absolutely no control over whether we live or die; and we know it. This might seem obsessive, but the thinking must go that if a reader follows these disasters often enough by rereading the CVR transcripts, when (and if) the time comes to experience one such incident for real he or she will be ready. Maybe that is true for some people. They have already been there, so to speak. But I also suspect that more readers follow the transcripts for their drama, as I did at the start. It is undeniable that they make riveting reading, because they document real life-and-death events as they unfold minute by minute from a spectacular angle. We can follow the activities, emotions and voices of cockpit crews from the instant something goes awry to a final outcome. And what can be more dramatic than for-real death or salvation? All drama, whether portrayed as fiction or fact, is necessarily voyeuristic. And what can be more intrusive than peeking from behind the curtain at the last frenzied, intimate moments in another human being’s life?

Unsafety will be with us in the air for a long time to come. Flying in North America and Europe may have reached a point of statistical perfection but we will still have the Third World, which is where airlines are crashing today.

The imbalance in safety between different parts of the world is stark.

In March 2007, Russian Airlines UTair Flight 471, a Tupolev Tu-134, which crashed while attempting to land at Samara’s Kurumoch Airport, in Russia, killing six of the fifty-seven passengers on board, was only one of two fatal commercial passenger aircraft accidents that did not occur in the Third World or involve an aircraft registered there. (The worst aviation disaster of 2007 was the crash of the Brazilian TAM Linhas Aeéreas Flight 3054, an Airbus A320 that overran the runway at Congonhas-Saão Paulo International Airport in Brazil, killing 187 on board and 12 on the ground.) In the Samara incident the aircraft was a Tupolev; in case you did not already know, boarding any Tupolev anywhere, flown by any airline, whatever its destination, is guaranteed to be the thrill of a lifetime. My wife and I flew in one a few years ago from the Bahamas to Havana, Cuba. The subsequent vacation, the cigars, music, food and sun and rum were just a pleasant afterthought to the joy of having landed alive.

As in so many other aspects, Africa, in terms of air safety, has become a special case, with the European Union banning most non-national African airlines from landing at EU members’ airports. Last year the Congo saw more fatal commercial air crashes than any other country. Four cargo aircraft and two Let 41 passenger flights suffered fatal crashes. One of these accidents involved an Africa One plane which came down in Kinshasa, killing so many people on the ground that no precise number of casualties was ever given. The downward trend in the safety of African airlines is long and continuous. Even as far back as the early 1970s, flying in Africa required a cavalier attitude. I remember when, based in Kenya with Newsweek magazine in 1973, I was aboard an Air Zaire flight from Kinshasa to Nairobi, and, to my surprise, a beautiful young woman who had boarded the Boeing 707 aircraft in Kinshasa simply vanished soon after takeoff. I know because I looked for her. A week later I ran into her at a Nairobi dinner party. She laughed as she recalled the thrill of the flight, telling me that, although unqualified, she had piloted long stretches between Bujumbura and Entebbe while sitting in the captain’s lap swilling goblets of champagne.

To change this direction in African air safety, the American NTSB and others are working ‘aggressively’ to help African nations. At first glance the continent would appear to extend beyond the NTSB’s mandate, and indeed it does…and yet doesn’t. African governments and private airlines based in Africa buy and fly American-made aeroplanes and helicopters, which gives the NTSB an inherent interest. ‘For commercial purposes, we don’t want them crashing Boeings’, a member of the NTSB told me. ‘We want them buying Boeings. If there are crashes, the African governments or airlines might say that the plane was no good. “Next time we’ll buy Airbus”.’ This was the line the Egyptian government took when one of EgyptAir’s pilots committed suicide in October 1999 by crashing a Boeing 767 into the Atlantic Ocean. The Egyptian government alleged that the 767’s flap system was to blame. It wasn’t. ‘That was not good for business,’ the same NTSB representative told me.

Air safety in Africa can also have serious political ramifications. In August 2005, John Garang, the newly sworn in vice president of the Sudan, was killed when his helicopter crashed during an official trip to Uganda. Soon after, the BBC reported ‘large-scale’ rioting in the Sudanese capital Khartoum, with supporters of Mr Garang battling armed police. They inferred from the news that Sudanese enemies in positions of authority had ordered his killing. This suspicion was perhaps inspired by the shooting down of a plane in 1994 that was carrying Rwanda’s President Habyarimana, an incident that served as a flashpoint for the subsequent genocide. Immediately after the 2005 crash in Uganda, the US Department of State dispatched one of the NTSB’s seasoned investigators, Dennis Jones, to the crash scene. His conclusion that foul play was not involved may have prevented a bloody civil war.

For readers unfamiliar with CVR transcripts, an explanation is in order. The transcripts are taken from recordings of sounds of interest to investigators after crashes. These are inter-cockpit voices, engine noises, stall warnings, landing gear extension and retraction, and all sorts of other clicks and pops. Investigators can often determine from these noises parameters such as engine rpm, systems failures, speed, and the time at which certain events occur. The CVR tapes also record communications with air traffic control, automated radio weather briefings and conversation between the pilots and ground or cabin crew.

All over the world these recordings are contained in boxes carried in the parts of commercial aircraft most likely to survive a crash, such as the tails. These boxes are known colloquially as black boxes; there is one for cockpit voice recordings (CVR) and another for flight data recordings (FDR). In the cockpit, the crews’ voices and other sounds are detected by ‘cockpit area microphones’,or CAM, usually located on the overhead instrument panel between the two pilots. The older analogue CVR units use a quarter-inch magnetic tape as a storage medium on a thirty-minute self-erasing loop. Newer models use digital technology and memory chips for up to two hours of self-erasing recordings. The boxes contain an underwater locator beacon (ULB), which activates a ‘pinger’ when the recorder is submerged in water and transmits an acoustical signal on 37.5 KHz that a special receiver can detect at depths of 14,000 feet. The boxes can sustain a crushing impact of 3400 gravities of force. One even sustained 9000 gravities after the crash in 1987 of a hijacked Pacific South-west Airlines flight in California.

The cockpit voice recorder (CVR)