Sure, Pilots Hate 'Em, But Some Think They're Vital

by ANN Contributor Steven R. Lund

Sure, the FAA wants better audio recording capabilities on cockpit voice recorders -- a move the agency says would bring the aviation industry more in line with current documentation technology. But what about video?

In July 2004 the US National Transportation Safety Board conducted a formal hearing to scrutinize the feasibility, potential benefits and drawbacks of requiring crash-protected cockpit imaging systems, also called video recorders.

Over the course of the hearing's two days the NTSB heard testimony from regulators, manufacturers, investigators, and pilots to synergistically consider the issue from various points of view. Some pertinent subjects examined at this hearing included the following:

• The applicability of technologies like data encryption and how this technology may be used to protect and secure images (and other) data the technical feasibility and cost associated with capturing, storing, and analyzing image data obtained from commercial aircraft cockpits
• The potential accident investigation benefits of cockpit image recorders (CIRs)
• The legal and privacy concerns of flight crews, air carriers, regulatory agencies and international organizations
• The regulatory difficulties surrounding implementation of image recorders

The applicability of technologies to record pertinent images has been thoroughly examined over the last five years.  The Safety Board has conducted several symposia where individuals from industry, unions, and government made presentations pertaining to the issues of cockpit image recorders. One of these, was in May 1999-The International Symposium on Transportation Recorders, which was cosponsored by the Safety Board and the International Transportation Safety Association. This symposium prompted the FAA to form a committee composed of industry, labor, and government experts to examine emerging flight data recorder technology, including cockpit image recording devices.

Witnesses testified at last year's hearing that a video recording would have helped solve the small plane crashes that killed Sen. Paul Wellstone of Minnesota in October 2002 and members of the Oklahoma State University basketball team in January 2001.

The potential benefits of Cockpit Image Recorders to accident investigation was conclusively proven during the rejected takeoff accident investigation involving an Overseas National Airways (ONA) wide body aircraft that went off the wet runway at New York's JFK airport in November 1975. Case in point:

• The plane ingested numerous large seagulls into the right wing engine just before takeoff prompting the pilot to initiate an abort in an unsuccessful attempt to stop his high takeoff gross weight airplane on the wet runway. The RTO was not successful because of several factors:
• The engine caught fire, began breaking up and depositing hot debris onto the runway directly in the path of the main landing gear tires
• At least two, possibly three, of the four main landing gear tires failed, severely impeding the braking on the right main gear
• The hydraulic line to the brakes was severed on this damaged gear, virtually eliminating any braking ability on that gear and depleting that hydraulic system of its fluid
• the lack of braking ability and the asymmetry caused by the uneven braking, asymmetric reverse thrust and asymmetric ground spoilers (due to loss of one of the three hydraulic systems powering the spoilers) caused the aircraft to uncontrollably drift off to the right and off the runway
• The damaged landing gear collapsed upon exiting the hard surface, the fire spread to the engine pylon and wing and the aircraft was destroyed

Fortunately, the passengers were trained flight crew members being repositioned for a foreign operation and all escaped the burning plane with no serious injuries.

Objective Use Of Cockpit Video In Accident Investigation

The ONA accident proved the virtue of cockpit video by the fortuitous fact that an off duty captain, seated in a cockpit jump seat behind the pilot in command was using a hand held movie camera mostly trained on the instrument panel with the throttle quadrant and pedestal in the foreground during the takeoff roll and up until the plane exited the hard surface.

Despite the fact that the camera was hand-held, hence not fixed to any particular constant field of view, investigators were still able to do a frame by frame analysis revealing simultaneous events to objectively correlate the cockpit situation with data provided by CVR and DFDR information, clearly showing the entire sequence of events leading to the accident, including:

The rapid pilot reaction time when the "ENG. FAIL" light illuminated as the seagulls severely damaged the engine fan section.

The airspeed indicator the pilot would be monitoring showing the speed at which a safe abort should have been possible when the birds hit the fan.

The comprehensive, objective information from the cockpit images were then correlated with the sounds on the Cockpit Voice Recorder (CVR) and all the data on the Digital Flight Data Recorder (DFDR) with its very accurate time information.

An accurate assessment of the pilot reaction time compared to FAA certification standards for pilot reactions for the anticipated case of an engine failure on takeoff, which was designed to ensure a safe stop within the stopping distance demonstrated to the FAA during the certification process for all commercial aircraft

This objective cockpit image data also substantiates the statement made by the witness at the NTSB hearing from the British Aircraft Accident Investigation Branch [AAIB], the NTSB's counterpart in the UK, who testified "that there might be gaps left between the DFDR and CVR data." The dialogue recorded on the CVR between the pilots "often leaves much to be desired."

"It's common to hear on CVR recordings 'Look at this' and we sit there wondering what 'this' is," the head of the AAIB testified at the hearing.

Investigators have been frustrated following numerous investigations where Cockpit Voice recordings were totally lost because of the Bulk ERASE feature designed into the current systems ostensibly to protect Pilots' privacy-a system that provides a button in the cockpit that will electrically degauss the entire CVR tape when pushed. This system was sometimes activated by electrical transients during the crash, rendering the tape useless. 

The highly accurate, objective sequence of events the investigators were able to document on the ONA accident coupled with other high energy RTO accident investigations compelled the industry to completely reconsider the RTO maneuver. From the FAA Certification criteria of empirical stopping distance determination, theoretical calculation using a standard set of pilot reaction times, and aircraft hardware changes providing additional information to the flight crews, including:

• Display all the M.L.G. tire/wheel temperatures and pressures
• Adding a tire failure aural warning to enable pilots to distinguish a tire failure from the "bang!" associated with a failure of an engine
• Adding a system that would automatically apply the required braking for a given situation, including the RTO mode, changing the brakes from steel to a higher energy absorbing carbon material
• An improved anti-skid system under digital control with self testing capability before every takeoff.

The FAA changed the certification test standards to require high energy takeoff acceleration then deceleration to a full-stop to convincingly demonstrate an RTO can be safely performed within the allotted stopping distance for each airplane. This certification substantiation was previously made during a landing roll by applying full anti skid braking when the plane had decelerated to the calculated speed for the weight being tested to simulated a high takeoff weight rejected takeoff using brand new tires and cold brakes. While this test showed that the brakes could absorb the requisite energy for the weight being tested, it apparently did not realistic represent the actual RTO condition in service where:

• The airplane was accelerating during takeoff rather than decelerating when the brakes are first applied
• Aircraft in-service virtually never operated with new tires and brakes
• the brakes on the in-service airliners might have some residual heat from being applied during taxi
• the takeoff might be rejected due to a tire failure, hence severely limiting stopping capability.

These facts became evident following the investigation of several high energy RTO accident investigations, firmly reinforced by the cockpit images recorded during the ONA accident, then were substantiated by follow-up simulator exercises and taxi tests on actual aircraft. In the absence of Cockpit Image information, investigators concluded that the most probable cause of the failure of the aircraft to stop on the runway was the pilot failure to react in a timely manner. But, the ONA cockpit images firmly proved this was NOT the case. Hence, the further industry studies producing radical changes in the national aviation system to prevent further RTO accidents.

The manufacturer of the plane involved in the ONA, and other RTO accidents, invited commercial pilots to participate in a simulator training exercise designed to measure flight crew responses during takeoff with simulated failures programmed to mimic those experienced in actual RTO accidents. The engineering simulator used was instrumented to record the same parameters as those recorded on the DFDR, plus the cockpit environment was recorded by video during each simulated run. The results of these similar tests showed some surprising results:

The pilot's-in-command tended to reduce the full braking effort once the plane's speed slowed, resulting in a simulated runway overrun as in some accident cases.

Some pilots logically thought that what the FAA had euphemistically dubbed "the takeoff decision speed" or V1 speed was the highest speed they would allow before beginning to stop during a rejected takeoff, when the actual case of making the decision AND accomplishing all necessary actions to stop the plane before reaching the V1 speed is the case, because of the manner in which the stopping distance was predicated during certification: of landing the plane beginning the stop with the aerodynamic drag inducing ground spoilers deployed and waiting for the testing V1 speed then applying full brakes to bring the test plane to a full stop in the measured distance.

Those pilots who waited until reaching the V1 speed before deciding to initiate the aborted takeoff in the simulator after a simulated failure prompting an abort, invariably went off the simulated runway at a speed directly proportional to the speed above the V1 speed the abort was initiated.

Theoretical calculations and these simulator tests showed that if pilots attempted an aborted takeoff above the V1 speed, the question became, how fast the plane would be going when it went off the end of the runway, rather than if it would stop on the runway.

The tests conducted in an actual airplane with instrumented tire pressure and temperatures during taxi showed that excessive taxi speeds and distances tended to heat the wheels, tires, and brakes to unacceptable levels that would not permit a safe stop in the event of a high energy RTO after the taxi.

When considering the unprecedented flight safety enhancements made to the national aviation system (NAS) stemming from the direct result of only one accident investigation using the objective information investigators obtained from one isolated, hand-held, cockpit image recorder. This leads one to wonder what the vast impact would be to improving aviation safety overall should Cockpit Image Recorders be required in all airplanes to be used in the investigation of aviation mishaps.

In 2000 the Board addressed the larger category aircraft. Recommendation A-00-30 asked the FAA to require the retrofit of existing aircraft operated under 14 CFR Parts 121, 125, or 135, and currently required to be equipped with a CVR and digital flight data recorder (DFDR) with a crash-protected image recorder system. Safety Recommendation A-00-31 was identical, except it covered newly manufactured aircraft.

Cockpit Video Recordings For Flight Crew Training

In addition to aiding investigators in objectively determining the most probable cause of accidents and incidents, CIRs have, and will continue to, act as a valuable flight training aid to greatly enhance future flight safety. For instance, the images from the ONA RTO at JFK were used to make an RTO training Video currently being used by airlines world wide. After the aircraft manufacturer of the plane involved in the ONA accident produced the training video and distributed it free of charge to several customer airlines throughout the world, one foreign airline was given the video script to reproduce the video with foreign subtitles. In addition, another manufacturer of large jetliners also produced a similar training video.

A valuable tool in the simulator training in Cockpit Resource Management by a leading US Airline was a cockpit video recording of flight crew responses to simulated emergencies, which was later analyzed by the crew in training to realistically show their performance. This very successful training program was initiated with the proviso that the video recordings would be destroyed immediately following completion of the training.

Legal Issues

To protect the privacy of pilots, CIRs in actual airline operation, would be covered by the same law that currently protects Cockpit Voice Recordings, but the law would have to be strengthened to include foreign operations as indicated by the words of the Airline Pilots Association's (ALPA) executive safety committee chairman, who expressed serious concerns about keeping recordings from the media once the plane enters international airspace, pointing to the voice recording from an American Airlines jet that crashed in Colombia in 1995: "parts of that tape were broadcast by NBC," said the ALPA spokesman.

But, surely when legislators weigh the life-saving benefits CIRs provide against the pilot privacy issues, stern international statutes will be enacted to protect pilots from improper public disclosures of CIR information, to be used in litigation. For example, just as NTSB reports currently are not admissible in court under US law, and as the existing International Civil Aviation Organization (ICAO) rules stipulate. So, when the CIR information is used for accident investigation only and is kept strictly confidential, there will be no question about the life-saving benefits Cockpit video will provide aircraft accident investigators in their task of preventing future accidents.

Attempts to mandate such a valuable asset for accident investigators would naturally be resisted by the pilots and their Airline employers because such counterproductive, unauthorized use of CIR by the media would prompt Plaintiff's Attorneys to undertake frivolous legal action against the pilots in the recording because of ignorance and misinterpretation of the cockpit situation. So, unauthorized public release of this recorded information must be strictly prohibited by international Laws before any Cockpit recordings (Video or Voice) can be used effectively to radically improve future Flight Safety.

(Steve Lund is the retired Director of Flight Safety Investigations from Douglas
Aircraft Co. -- now Boeing -- in Long Beach, CA)

FMI: www.ntsb.gov