One of the most scrutinized pieces of evidence gathered from an aircraft accident is the collection of information contained in the Cockpit Voice Recorder and Flight Data Recorder. CVRs and FDRs paint an often haunting, but frequently useful picture of what occurred during the last minutes of an accident flight. This is not to say, however, that the recorders are always conclusive, or even useful. There are a handful of cases where the CVR and FDR tapes have broken, failed to record, stopped recording early, or not captured enough information to be useful to the investigation.
Advancements in these devices are not new to the industry; however, the pace is slow to incorporate new technology into current fleets. Justification Concerns surrounding Cockpit Voice Recorders and Flight Data Recorders stem from all corners of the aviation industry. There are currently five outstanding NTSB recommendations to the FAA regarding the use of and reliability of CVRs and FDRs. Of these five suggested improvement areas, the FAA has yet to respond to any of them, prompting the NTSB to place the issues on their list of Top 10 Most Wanted Safety Recommendations.
Documented cases will be presented in this text where CVRs and FDRs have stopped recording seconds, and even minutes, before an accident. Other cases will examine incidents where the focus shifts to a hypothesis as to what may have been recorded on earlier portions of the tape. Either way, strong cases will be built to justify having CVR and FDR tapes not only record longer, but to record more information as well. Herein, we shall identify existing problem areas, areas where work is ongoing, and areas for which future plans are in existence. These topic areas can be identified as follows:
FDR and CVR carriage requirements for new aircraft FDR and CVR carriage requirements for existing aircraft (retrofit) Independent power supplies Cockpit video recording Deployable recorders NTSB recommendations Background History Flight information recorders have been in use on commercial aircraft since the 1950s. The FAA requires both CVRs and FDRs to be installed on all aircraft capable of carrying ten or more passengers when used in scheduled revenue service (FRD/CVR). The recorders, also referred to as black boxes, are installed to help reconstruct the events leading up to an aircraft accident.
There is no disputing the fact that black boxes are essential components of civil aircraft today; their job is critical to the investigation of an accident, however, safety experts would argue that the basic capabilities of these devices are no longer sufficient. Specifications The International Civil Aviation Organization (ICAO) is a regulatory body, which sets standards for international aviation. ICAO is broken into 18 annexes, and, comparable to international law, they are the very basic requirements that the organizations 185 member states must follow.
Annex 6 to ICAO deals with the operation of aircraft and states that CVRs should be capable of retaining the information recorded during the last 30 minutes of the devices operation. In addition, ICAO has recommended that CVRs installed in aircraft over 5,700kg with a certificate of airworthiness issued after January 1, 1990 be capable of retaining information recorded during the previous two hours of operation. ICAO suggests that flight recorders be constructed, located and installed so as to provide maximum practical protection for the recording in order that the recording information may be preserved, recovered and transcribed.
The NTSB has submitted five recommendations to the FAA regarding the reliability of recorders in commercial and civil aviation aircraft. The recommendations are based on historical data stemming from a series of accidents where forces interfered with the recording, and inaccurate or incomplete data was recovered from the scene of an accident. A partial list of accidents will help in evaluating the need for improved recording devices, however these are only the most recent cases in a long history of accidents where the investigation was hindered by a loss of data (Safety Issue).
May 11, 1996, ValuJet 592, a DC-9-32, crashed shortly after departing Miami, Florida. The recorders stopped recording about 40-50 seconds before impact. All 111 onboard were killed. July 7, 1996, TWA 800, a Boeing 747-100, exploded 13 minutes after takeoff from New York enroute to Paris. The recording ended at the time of the explosion, however, the craft remained airborne for another 40-50 seconds. All 230 onboard died. December 19, 1997, Silk Air 185, a Boeing 737, entered a high-speed descent from 35,000 feet and impacted terrain near Palembang, Indonesia.
The investigation revealed that both recorders stopped recording before the vessel entered the rapid descent. There were 104 fatalities. September 2, 1998, Swissair 111, an MD-11, crashed into the waters off Peggys Cove, Nova Scotia. The investigation is hampered by the lack of data from the recorders, which stopped 6 minutes before impact. All 229 aboard were killed. Since 1983 , there have been 52 accidents and incidents where information from the CVR or FDR were lost due to the interruption of electrical power (Safety Issue).
There have been 23 documented cases in which CVRs and FDRs were destroyed by impact forces, fire, or from a failure in the structural integrity of the box (Aviation Safety Network). Discussion and findings Carriage requirements For new aircraft ICAO proposes that flight data recorders for aircraft for which the certificate of airworthiness is issued after January 1, 2003, and whose empty weight exceeds 5,700 kg, be required to be fitted with a Type IA FDR. Type IA FDRs record the parameters required to determine accurately the aircrafts flight path, speed, attitude, engine power, configuration and operation.
These parameters would only be recorded if the data were readily accessible on the aircraft, something most new aircraft are already capable of complying with. ICAO also proposes that CVRs installed in aircraft over 5,700 kg and certificated after January 1, 2003 be capable of retaining the information recorded during the previous two hours of operation (Frostell 20). To date, neither ICAO, nor the FAA have been successful in influencing any mandates for carriage requirements on new aircraft. For existing aircraft Aircraft certificated after January 1, 1989 and weighing 27,000 kg or more must be equipped with a Type I FDR.
For aircraft certificated at the same time, but weighing between 5,700 and 27,000 kg, Type II FDRs are required for operation. Type I and Type II flight data recorders have a duration of 25 hours, the difference being that Type I recorders monitor 32 parameters, whereas the Type II monitors 15. All aircraft certificated after January 1, 1987 and weighing more than 5,700kg are required to be equipped with a CVR which monitors the aural environment of the flight deck during all phases of flight. The CVR must be capable of recording information from the last 30 minutes of its operation (Frostell 20).
Independent power supplies Specifically stemming from the crash of Swissair 111, in which both flight recorders stopped collecting data six minutes before the accident, the FAA and Transportation Safety Board of Canada recommend that recorders be fitted with independent power supplies to provide alternate current to recorders in the event of a loss of power. Redundancy in aviation is a necessity, and in most aircraft flying today, there are multiple components that perform the same function, each one backing up the other.
Because there have been accidents involving loss of power, and because recorders have been known to cut out during losses and surges, having an alternate electrical source seems logical. NTSB recommendation A-99-16 is currently open and awaiting response from the FAA. If adopted, it would require a retrofit after 1 January 2005 on all airplanes required to carry both a CVR and FDR to be fitted with an independent power source located with the CVR that would provide 10 minutes of operation after automatically engaging when aircraft power to the recorder is interrupted (Safety Issue).
Recommendation A-99-18 expands on the earlier recommendation to require that CVRs and FDRs be powered from separate generator buses with the highest reliability. This would allow at least one recorder to remain useful in the event that one bus fails and electricity to the unit is cut off. For an FDR, an independent power source would not, however, necessarily enable the recording of any additional information. Backup power to an FDR would result in the continued recording of data from sensors that are still powered, and can relay that information to the FDR.
If all aircraft power is lost, the probability is small that the sensors and unit would be receiving power or be able to relay the information to the FDR. In the case of the CVR, the microphone is powered by the acquisition unit. As long as the CVR is powered independently, the microphone will continue to operate, and the CVR will continue to record sounds regardless of whether the aircraft is powered. The following list contains only some of the more notable accidents for which vital flight recorder information was not available because of a premature loss or interruption of electrical power.
Independent power sources may be instrumental in providing information to investigators when accidents like these occur (Safety Issue). November 12, 1975, Overseas National Airways 032, a DC-10-30 on takeoff roll from New York JFK Airport ingested a flock of seagulls. The right engine disintegrated, causing a loss of power to the FDR. The airplane and CVR were destroyed by postcrash fire. There were no fatalities. April 4, 1977, Southern Airways 242, a DC-9-31 penetrated a level 6 thunderstorm over New Hope, GA, causing the failure of both engines.
Both recorders stopped following the engine failures and did not resume operation until 2 minutes 4 seconds later when the APU was started. The aircraft crashed on a two-lane highway and of the 85 people onboard, only 23 survived. May 25, 1979, American flight 191, a DC-10-10 on takeoff rotation from Chicago OHare experienced an engine separation following an engine failure. Power to the FDR sensor and CVR was lost. The airplane crashed 32 seconds later. All 271 onboard and 2 persons on the ground were killed.
February 24, 1989, United flight 811, a Boeing 747-122 experienced a forward cargo door separation over Honolulu, Hawaii. As a result, the number 3 and 4 engines were shut down because of foreign object damage. The CVR was lost for 21. 4 seconds during the emergency descent as a result of the shutdown. Nine people onboard died. Cockpit video recording The idea behind recording images on the flight deck had its roots planted long before the crash of Egypt Air flight 990. But the mystery surrounding that accident in particular, makes a strong case for having visual evidence to back up that which can be heard in the cockpit.
Not much has been said about the Egypt Air disaster lately, but a cloud of suspicion hovers around the idea of foul play. Flight 990, a scheduled passenger flight from New York to Cairo, ditched into the waters off Nantucket on the night of October 31, 1999. When the black boxes were recovered, the only clues to what could possibly have occurred were voices and the sounds of buttons being pushed in the cockpit. Investigators are unsure about who entered the flight deck, and exactly what went on in there. Cameras could have helped to solve the mystery.
A company called Sight Recorder has developed a video recording system that it claims could have solved every recent crash mystery by providing investigators with full color digital images. Sight Recorder allows the crew to view color images from mounted cameras in the tail, belly, cabin and cockpit in hopes that they might detect irregularities before they become critical (Sight Recorder). With clear color images, Sight Recorder hopes to leave no mystery of causation to frustrate investigators after a crash. Deployable recording devices The United States military has been using deployable recorders on certain aircraft for years.
The devices, with frangible attachments, are designed to pop free of the aircraft on impact or when immersed in water, and separate itself from the rest of the wreckage (Nordwall 86). With the latest rash of airline accidents occurring over and in water, it seems logical to have a recorder that would eject itself and float on the surface. Floating recorders may also mitigate the monetary and time costs for surface vessels and divers to recover the boxes, and allow a more concentrated effort to be put toward rescue and recovery of bodies and crucial wreckage.
The DRS Company has developed a combination CVR/FDR with an emergency locator transmitter (ELT) in a single package. When ejected, the ELT is activated and the recorder drifts to the surface, landing softly and floating indefinitely. For the time being, deployable recorders are only being used on military aircraft, but if found viable and valuable, may be a worthy investment in civil aviation. NTSB recommendations A-99-17 The NTSB issued Recommendations A-99-16, -17, -18 to the FAA on March 9, 1999.
All three recommendations remain open and the NTSB has heard no word from the FAA on any plans to address them. Recommendation A-99-16 and 18 were outlined in the discussion on independent power supplies. A-99-17, however, deals with dual recording systems. If adopted, the recommendation would require all aircraft manufactured after January 1, 2003 that are required to carry both a CVR and FDR to be equipped with two combination CVR/FDR recording systems. One system would be located as close as possible to the cockpit as practicable and the other as far rear as possible.
Both recorders should have the capability to monitor all the required parameters covering the previous 25 hours of operation, and all cockpit audio for the previous 2 hours of operation. The system near the cockpit should be provided with an independent power source, capable of recording 10 minutes of audio whenever normal aircraft power ceases. The bus that provides the greatest amount of reliability for operations without jeopardizing service to essential loads should power the aft system, whereas the bus that provides the second greatest reliability should power the system near the cockpit (Safety Issue).
This recommendation stems from the accidents involving TWA 800 and ValuJet 592. In both cases, the CVR and FDR were located near the rear of the aircraft for greater survivability. Rear mounting usually results in long cable runs from the cockpit to the recorders, and in these cases, the wiring between the bus and recorder had been severed or the signal compromised. In either case, having two independent recording systems would eliminate the vulnerability of the signal wires to damage. Conclusions and Recommendations Too often, the voice of the NTSB is muffled, shut up by the boisterous bellows of Congress, the airlines, unions and the FAA.
With the issue of CVRs and FDRs, I believe that commercial aviation can no longer stand idly by and wait for another accident, only to reveal another useless tape. As evidenced by so many recent accidents, trouble in the cockpit can strike at any time, and last for any duration. The industry can benefit from recordings that begin earlier and capture more data. Thirty-minute CVR tapes are insufficient, and even with a 10-minute independent power supply, a loss of power would result in 1/3 of the transcript being taped over.
A good case is presented for 2 hour-long tapes. As technology speeds into the blue yonder, we can expect to see more innovations in in-flight recordings. Civil aviation has not progressed to the point where video cameras and deployable recorders are viable and reliable devices. Will it ever? We can speculate. For now, the recommendations that lie before the FAA appear sufficient. The current trends and problems surrounding recorders have seemed to be addressed by the Safety Board. It is up to the Feds now.
