Aviation Investigation Report A98H0003
4.2.5 Accident Investigation Issues
- 220.127.116.11 - Quality of CVR Recording
- 18.104.22.168 - Quick Access Recorder Data
- 22.214.171.124 - Image (Video) Recording
Frequently, the CVR recording of cockpit conversations are of poor quality, particularly when the conversations are recorded through the CAM. The voice quality on CVR recordings is dramatically improved when voices are recorded through boom microphones. However, pilots are not required to wear headsets with boom microphones at cruising altitudes.
Various national regulations differ concerning the maximum altitudes below which flight crews are required to wear boom microphones. For example, the CARs require the use of boom microphones below 10 000 feet, the FARs below 18 000 feet and the Joint Aviation Requirements do not have any requirement that they be used. Swissair required their pilots to use boom microphones when flying below 15 000 feet. The present requirements were developed before modern technology allowed headsets with boom microphones to be designed for comfort over long periods of time, such as during cruise flight.
When the SR 111 pilots first noted an odour in the cockpit, they were in cruise flight and were not wearing boom microphones. Although the internal communications between the pilots were recorded through the CAM, the conversations were difficult to hear and decipher. There was a marked improvement in recording quality after the pilots donned their oxygen masks, which have built-in microphones.
Even though the boom or oxygen mask microphones are recorded on a different channel than the CAM, the recordings of internal communications on the microphone channels are still frequently masked by incoming radio transmissions because internal, as well as external, communications are recorded on the same CVR channels but at different amplitudes. For example, the recorded incoming radio communications for SR 111 were of significantly higher amplitude than the internal communication from the mask microphone, making it difficult and occasionally impossible to discern internal communications. The relative amplitude of the incoming radio calls to that of the internal communications is pre-set at equipment installation and is not affected by crew adjustment of audio volume. Therefore, even if the pilots can hear each other readily through their headsets, the CVR recording of internal communications may be masked substantially by incoming radio communications. Significant difficulties in extracting such "masked" internal communications from CVR recordings have been experienced by the TSB and by safety investigation agencies from other nations.
The ability to decipher internal conversations between flight crew members is an important element of effective accident investigation. Therefore, the Board recommends that
Regulatory authorities, in concert with the aviation industry, take measures to enhance the quality and intelligibility of CVR recordings.
Assessment/Reassessment Rating: Satisfactory in Part
Quick access recorders (QAR) are voluntarily installed in many transport aircraft and routinely record far more data than the mandatory FDR. For example, the FDR installed on SR 111 was a solid state unit that recorded approximately 250 parameters, whereas the QAR used a tape-based cartridge, which recorded approximately 1 500 parameters. That is, the optional QAR recorded six times the amount of data recorded on the mandatory FDR. The additional data recorded on the QAR included numerous inputs from line replaceable units (LRU) that would have been extremely valuable in determining aircraft systems status, as well as temperatures at a number of locations in the fire-damaged area.
Many airlines are developing Flight Operational Quality Assurance (FOQA) or Flight Data Monitoring (FDM) programs; such programs require that increased data sets be recorded. The use of QARs is voluntary; therefore, the operating environment allows operators to change the QAR data-set according to their operational requirements. Conversely, changing the data-set on an FDR is currently an expensive process, largely due to the associated re-certification issues. As modern-day versions of both types of recorders employ solid state memory technologies, these modern FDRs effectively have as much capacity to record data as QARs. The Board believes that there is no technical reason why safety investigations should not benefit from the FOQA/FDM trend, and that all data voluntarily collected for any operational purpose should also be available for accident investigation. To achieve this, regulatory authorities need to develop regulations that protect the core parameters required for all FDRs, while also allowing FDRs to be easily augmented with additional parameters, higher sample rates, and higher resolutions without requiring re-certification of the FDR and without requiring validation/calibration of parameters that are not dedicated to the FDR. Operators would need ready access to these FOQA/FDM parameters and might choose to use only the FDR unit to meet the mandatory FDR parameter list, as well as their optional FOQA/FDM data needs.
The Board recognizes that the US convened a Future Flight Data Collection Committee to address these issues, and that in Europe, the European Organisation for Civil Aviation Equipment (EUROCAE) Working Group 50 is updating its international Minimum Operational Performance Specifications. The Board supports FOQA and FDM programs and believes that they contribute significantly toward improving aviation safety. The Board also believes that all FOQA and FDM data routinely collected should be available for safety investigations. Therefore, the Board recommends that:
Regulatory authorities require, for all aircraft manufactured after 1 January 2007 which require an FDR, that in addition to the existing minimum mandatory parameter lists for FDRs, all optional flight data collected for non-mandatory programs such as FOQA/FDM, be recorded on the FDR.
Assessment/Reassessment Rating: Unsatisfactory
Only recently has it become economically feasible to record cockpit images in a crash-protected memory device. New "immersive" technology provides for camera systems that can capture panoramic, wide-angle views necessary to record the cockpit environment. Image recordings can capture other aspects of the cockpit environment that would otherwise be impractical or impossible to record. Special playback software allows investigators to "immerse" themselves in the cockpit and view virtually the entire flight deck.
Vital information regarding the cockpit environment, non-verbal crew communications, crew workload, instrument display selections and status have not been available on traditional data and voice recorders. This has limited the scope of many investigations, but more importantly, has hindered the identification of safety issues and consequently the corrective action needed to prevent future occurrences.
Some operators are installing video cameras for operational purposes. These systems provide the flight crew with images, such as the external views of the undercarriage area, wings and engines, or internal views of cargo and cabin areas. Since these video images have the potential to influence critical operational decisions, the images presented to the flight crew should be stored in crash-protected memory to facilitate safety investigations.
The Board believes that image recording in the cockpit will substantially benefit safety investigations. It will provide investigators with a reliable and objective means of expeditiously determining what happened. This will assist safety investigators in focusing on why events took the course they did, what risks exist in the system, and how best to eliminate those risks in the future.
The Board endorses the NTSB recommendations issued in April 2000 (A-00-30 and A-00-31), and advocates the development of international Minimum Operational Performance Specifications for image recording systems by EUROCAE Working Group 50. Therefore, the Board recommends that:
Regulatory authorities develop harmonized requirements to fit aircraft with image recording systems that would include imaging within the cockpit.
Assessment/Reassessment Rating: Satisfactory Intent
The Board is acutely aware of the concerns expressed by industry associations that sensitive recordings will be inappropriately released to the public or used for purposes other than safety investigation. While Canada treats these recordings as privileged, all nations do not. If image recordings are to be universally accepted, worldwide protections need to be put in place for all cockpit voice and image recordings. These protections would allow investigation authorities to use the recordings for safety purposes while preventing them from being aired for other purposes. Therefore, the Board recommends that:
Regulatory authorities harmonize international rules and processes for the protection of cockpit voice and image recordings used for safety investigations.
Assessment/Reassessment Rating: Fully Satisfactory
- Date modified :
- Date modified: