Skip Ribbon Commands
Skip to main content
Safety Recommendation Details

Safety Recommendation A-14-126
Details
Synopsis: On January 7, 2013, about 1021 eastern standard time, smoke was discovered by cleaning personnel in the aft cabin of a Japan Airlines (JAL) Boeing 787-8, JA829J, which was parked at a gate at General Edward Lawrence Logan International Airport (BOS), Boston, Massachusetts. About the same time, a maintenance manager in the cockpit observed that the auxiliary power unit (APU) had automatically shut down. Shortly afterward, a mechanic opened the aft electronic equipment bay and found heavy smoke coming from the lid of the APU battery case and a fire with two distinct flames at the electrical connector on the front of the case. None of the 183 passengers and 11 crewmembers were aboard the airplane at the time, and none of the maintenance or cleaning personnel aboard the airplane was injured. Aircraft rescue and firefighting personnel responded, and one firefighter received minor injuries. The airplane had arrived from Narita International Airport, Narita, Japan, as a regularly scheduled passenger flight operated as JAL flight 008 and conducted under the provisions of 14 Code of Federal Regulations (CFR) Part 129.
Recommendation: TO THE FEDERAL AVIATION ADMINISTRATION: Require Boeing to improve the quality of (1) the enhanced airborne flight recorder radio/hot microphone channels by using the maximum available dynamic range of the individual channels and (2) the cockpit area microphone airborne recordings by increasing the crew conversation signals over the ambient background noise.
Original recommendation transmittal letter: PDF
Overall Status: Open - Unacceptable Response
Mode: Aviation
Location: Boston, MA, United States
Is Reiterated: No
Is Hazmat: No
Is NPRM: No
Accident #: DCA13IA037
Accident Reports: ​Auxiliary Power Unit Battery Fire Japan Airlines Boeing 787-8, JA829J
Report #: AIR-14-01
Accident Date: 1/7/2013
Issue Date: 12/1/2014
Date Closed:
Addressee(s) and Addressee Status: FAA (Open - Unacceptable Response)
Keyword(s):

Safety Recommendation History
From: NTSB
To: FAA
Date: 2/26/2019
Response: Our investigation of the Japan Airlines incident at Logan Airport found that the quality of the audio recording obtained from both EAFRs was poor. We found the following three problems. 1) The signal levels of the three radio/hot microphone channels were very low and used only about 25 percent of the available total dynamic range of the recorder. 2) Random, very short duration, full-deflection noise spikes could be heard throughout the recording. These random noise spikes used the full dynamic range of the radio/hot microphone channel recording. 3) During the airborne portion of the flight that was captured on the recording, almost all of the individual crew conversations on the cockpit area microphone (CAM) channel were completely obscured by the ambient cockpit noise. We issued Safety Recommendation A-14-126 based on these findings, which led us to conclude that the poor audio recording quality of the EAFR could impede future aircraft investigations, as the recorded conversations and other cockpit sounds might be obscured. We note that, along with Boeing, you assessed the audio recording quality by reevaluating the recordings from the Boeing 787 certification flight tests, and that your reevaluation confirmed the previous finding that audio from crew channels was clear and intelligible, as was the audio recorded by the CAM—including crew conversation and aural warnings. We further note that Boeing found no issues with the EAFR quality, and found the audio recordings to be of similar quality as recordings from other Boeing aircraft models. Boeing also evaluated the cockpit voice recorder (CVR) portion of the EAFR from the Japan Airlines incident at Logan Airport and concluded that all the channels were intelligible. Finally, we note that you agree with Boeing’s conclusions that the EAFR voice recordings were accurate, compliant, and acceptable. You told us that neither you nor Boeing had any reports of similar audio quality issues until you learned of the United Kingdom’s Air Accidents Investigation Branch’s (AAIB’s) concerns (discussed below). We understand that you and Boeing recently learned that most of the EAFR readers used by accident investigators had an obsolete version of the ground station audio decoding software furnished by General Electric (GE), and that this obsolete software version led to poor playback audio quality when decoding the data. You found that the audio quality improved when using a later version of the software. We note that operators who own GE EAFRs get updates of the ground station audio decoding software directly from GE, while government agencies do not receive the updates for free nor immediately learn when they are available. You indicated that there is no regulatory requirement for a recorder manufacturer to provide notifications or complimentary software updates to government agencies. Based on your assessment of the EAFR audio quality, you do not believe it is necessary to require Boeing to revise its 787 design, but we note that you plan to continue working with Boeing and with us to verify that the cause of the poor audio quality was the decoding software. In August 2017, our staffs met, along with staff from the AAIB and Boeing, in Renton, Washington, to discuss this recommendation. In July 2018, the AAIB released a report on its investigation of an April 29, 2017, serious incident involving a cabin depressurization that occurred on a Boeing 787-9, G-ZBKF, operating from London’s Heathrow Airport to New Delhi, India. The AAIB’s investigation revealed a number of audio quality issues in the incident airplane and in all Boeing 787 aircraft. These audio quality issues were discussed during the August 2017 meeting and were addressed in the AAIB report, which says that, when the AAIB played the EAFR audio recordings from the incident aircraft, the CVR recordings exhibited the same characteristics that led us to issue Safety Recommendation A-14-126, with approximately 10 percent of the available dynamic recording range being used when the crew headset microphones were in use. Subsequently, the AAIB became aware that the Australian Transport Safety Bureau and the Bureau d’Enquêtes et d’Analyses pour la Sécurité de l’Aviation Civile of France had also experienced the same issues when reviewing CVR recordings from Boeing 787 aircraft. We point out that this is in contrast to your assertion that there were no reported audio quality issues on the Boeing 787 other than what we found during our investigation of the Japan Airlines Boeing 787 8 incident at Logan Airport. The AAIB investigation found the same random, full-deflection noise spikes of very short duration that we found (item 2 from above) in our investigation, and contacted the recorder manufacturer for assistance, who then provided the updated version of the GE software. When the AAIB replayed the audio recordings using the updated software, the full volume clicks and pops were no longer present, confirming your finding. As discussed and demonstrated during our August 2017 meeting, the AAIB compared the Boeing 787 EAFR CVR recordings with a variety of other transport category turbofan aircraft, including the Boeing 747-400, 777, 767, 737-800, 737-300; Airbus A380, A340, A330, A320; and the Embraer 190. The results of the review indicated significant variation in the recorded dynamic range of the crew channels and the ambient background noise levels compared to crew speech recorded on the CAM channel. The lowest recorded dynamic range when the headset microphone was used was found in the Boeing 787, and the highest was in the Boeing 757. For the CAM channel, the Boeing 787 and Airbus A380 had the highest levels of background ambient noise, which, depending on the phase of flight, could mask crew conversation. Several CVR recordings were replayed at our August 2017 meeting, including the Boeing 787-8 EAFR CVR flight test recording. The intelligibility of crew communications recorded on the CAM channel during the flight test appeared clearer compared to the CAM recording from a routine flight. The AAIB report stated that the difference between the two recordings may have been due to the scripted nature of the flight test, resulting in the crew speaking louder and placing more emphasis on clarity than normal. However, evaluation of the Boeing 787 recordings also showed that when the aircraft cockpit’s heating, ventilating, and air conditioning (HVAC) system was turned off after flight, there was a notable reduction in the ambient noise recorded on the CAM channel, indicating that airflow from the HVAC system in the cockpit was contributing significantly to the high ambient noise on the CAM. It was also noted during our August meeting that the CAM recording gave the impression that the cockpit environment was noisier than and not representative of the actual ambient sound heard by someone listening in the cockpit during a flight. Although the AAIB’s investigation found issues similar to those that prompted us to issue Safety Recommendation A-14-126, the AAIB decided that it was not necessary to issue another recommendation for the same issue. We reviewed your actions and findings responding to three issues that led us to issue Safety Recommendation A-14-126. The following is a brief review of your responses. 1) You have not addressed the issue of the three radio/hot microphone channels’ signal levels being very low and using only about 25 percent of the available total dynamic range of the recorder. You believe this is not a problem, and that the EAFR audio recorder is compliant with certification requirements. Therefore, you do not believe any action is warranted to satisfy this part of the recommendation. 2) You have addressed the issue of the numerous, very short duration, random, full deflection noise spikes that used the full dynamic range of the radio/hot microphone channel. The technical problem that caused this issue has been identified and corrected through a software revision made by GE; however, although this addresses the problem, it appears that you cannot require that the software update be made available to accident investigation organizations. 3) You neither acknowledge nor plan to address the poor quality of the CAM channel audio recordings during the airborne portion of a flight. Based on the findings from our investigation of the Japan Airlines incident, discussions during our August 2017 meeting on this recommendation, and the results of the April 29, 2017, AAIB Boeing 787-9 investigation, we have determined that you have not addressed the issues that led us to issue Safety Recommendation A-14-126. We disagree that the audio from crew channels was clear and intelligible; nor was the audio recorded by the CAM, including crew conversation and aural warnings. Further, we disagree with Boeing that there were no issues with the quality of EAFR audio recordings and that they are of similar quality to recordings from other Boeing aircraft models. We also disagree with Boeing’s conclusion that the CVR portion of the EAFR from the Japan Airlines incident was intelligible on all channels. Pending your taking actions to address the three audio quality issues we found in our investigation, Safety Recommendation A-14-126 is classified OPEN--UNACCEPTABLE RESPONSE.

From: FAA
To: NTSB
Date: 10/22/2018
Response: -From Daniel K. Elwell, Acting Administrator: This is our final response to Safety Recommendation A-14-124 and in further response to Safety Recommendations A-14-125 and -126. These recommendations were issued by the Board on December 1, 2014, and this response supplements all previous letters. The Board issued these safety recommendations as a result of an incident that occurred on January 7, 2013, in which a Japan Airlines Boeing 787-8, JA8297, was parked at a gate at General Edward Lawrence Logan International Airport, Boston, Massachusetts, when maintenance personnel observed smoke coming from the lid of the auxiliary power unit battery case, as well as a fire with two distinct flames at the electrical connector on the front of the case. No passengers or crewmembers were aboard the airplane at the time, and none of the maintenance or cleaning personnel aboard the airplane were injured. In collaboration with the FAA, Boeing assessed the quality of EAFR audio recordings by re-evaluating the audio recordings from 787 certification flight tests. This re-evaluation verified that audio from crew channels was clear and intelligible, as was the audio recorded by the cockpit area microphone, including crew conversation and aural warnings. Boeing found no issues with the quality ofEAFR audio recordings and found them to be of similar quality to recordings from other Boeing aircraft models and cockpit voice recorder (CVR) manufacturers. Boeing also evaluated the CVR portion of the EAFR from the incident addressed by this safety recommendation and concluded that all the channels were intelligible. We agreed with Boeing's conclusions that the EAFR voice recordings were accurate, compliant, and acceptable. Boeing and the FAA did not have any reports of similar quality issues until we learned from the Board's specialists that foreign investigative agencies noted similar audio quality issues on the 787 EAFR. Boeing and the FAA recently learned that that most of the EAFR readers used by accident investigators had an obsolete version of the ground station audio decoding software furnished by General Electric (GE). This obsolete software version led to poor playback audio quality when decoding the data, and the audio quality improved when using a later software version. Operators who own GE EAFRs get updates of the ground station audio decoding software directly from GE, while government agencies do not receive them free of charge or immediately learn that one is available. There is no regulatory requirement for a recorder manufacturer to provide notifications or complimentary software updates to government agencies. Based on our assessment of the EAFR audio quality thus far, we do not believe it is necessary to require Boeing to revise their design. We will continue to work with Boeing and the Board to gather additional information addressing this concern to verify whether the cause of the poor audio quality was the decoding software. I believe that the FAA has effectively addressed Safety Recommendation A-14-124 and consider our actions complete. I will keep the Board informed of the FAA's progress on Safety Recommendations A-14-125 and -126 and provide an update by August 2019.

From: NTSB
To: FAA
Date: 9/27/2016
Response: We note that you and Boeing agree that the recording of stale data that we found in our investigation is not compliant with Part 25 standards, or with the requirements of Part 121, Appendix M, “Airplane Flight Recorder Specifications.” We further note that Boeing and its supplier investigated this issue and discovered that these deficiencies in recording flight and voice data were introduced after initial certification, when a replacement part for a component was introduced. Finally, we note that you are continuing to work with Boeing to develop and implement a corrective action plan to address these issues. We believe the corrective action plan must address resolving the specific components that caused this problem as well as why relevant FAA and Boeing procedures did not identify this issue before the replacement component part was introduced. Pending development and issuance of a corrective action plan that fully addresses the issue of stale data being recorded, including identifying and correcting the procedures that did not identify this problem before it was incorporated into the recorders on the aircraft, Safety Recommendations A-14-124, -125, and 126 remain classified OPEN—ACCEPTABLE RESPONSE.

From: FAA
To: NTSB
Date: 7/6/2016
Response: -Michael P. Huerta, Administrator: The FAA is continuing to work with Boeing to identify appropriate corrective actions to address A-14-124 and -126. Concerning A-14-125, the FAA and Boeing agree that the enhanced airborne flight recorder records stale data and is not compliant with certain Title 14, Code of Federal Regulations Part 25 standards, and thereby also does not meet the applicable requirements of Part 121, Appendix M, Airplane Flight Recorder Specifications. Boeing and its supplier further investigated and discovered that these deficiencies in recording flight and voice data were introduced after initial certification, when a replacement part for a component was introduced. The FAA will work with Boeing to implement a corrective action plan to address stale data. I will keep the Board informed of the FAA's progress on these recommendations and provide an update by July 31, 2017.

From: NTSB
To: FAA
Date: 4/16/2015
Response: We note that you are working with the Boeing Aircraft Company to investigate and address the enhanced airborne flight recorder issues in these recommendations. Pending completion of actions that satisfy each of the recommendations, Safety Recommendations A 14 124, -125, and -126 are classified OPEN—ACCEPTABLE RESPONSE.

From: FAA
To: NTSB
Date: 2/24/2015
Response: -From Michael P. Huerta, Administrator: The FAA is working with Boeing Aircraft Company to investigate and address the enhanced airborne flight recorder issues referenced in these recommendations.