驾驶舱语音记录器和飞行数据记录器联合下载报告

National Transportation Safety Board Office of Research and Engineering Washington, DC 20594

National Transportation Safety Board（美国国家运输安全委员会） Office of Research and Engineering（研究与工程办公室） Washington, DC 20594

DCA22WA102

DCA22WA102

COCKPIT VOICE AND FLIGHT DATA RECORDER

驾驶舱语音和飞行数据记录器

Combined Download Report July 1, 2022

合并下载报告 July 1, 2022

TABLE OF CONTENTS

目录

A. ACCIDENT...3 B. COCKPIT VOICE AND FLIGHT DATA RECORDER DATA RECOVERY GROUP...3 C. DETAILS OF THE INVESTIGATION...4 D. DATA RECOVERY...4 1.0 HFR5-V CVR DESCRIPTION...4 1.1 HFR5-V CVR Damage...5 1.2 HFR5-V CVR Recovery...5 1.3 Audio Recording Description...13 1.4 CVR Data Provided to CAAC...13 2.0 HFR5-D FDR DESCRIPTION...14 2.1 HFR5-D FDR Damage...14 2.2 HFR5-D FDR Recovery...15 2.3 FDR Data Provided to CAAC...23 2.3.1 FDR Plots...25 APPENDIX A. CVR QUALITY RATING SCALE...28 APPENDIX B. VALIDATED PARAMETERS...29

A. 事故...3 B. 驾驶舱语音和飞行数据记录器数据恢复小组...3 C. 调查详情...4 D. 数据恢复...4 1.0 HFR5-V CVR 说明...4 1.1 HFR5-V CVR 损坏...5 1.2 HFR5-V CVR 恢复...5 1.3 音频记录说明...13 1.4 提供给 CAAC 的 CVR 数据...13 2.0 HFR5-D FDR 说明...14 2.1 HFR5-D FDR 损坏...14 2.2 HFR5-D FDR 恢复...15 2.3 提供给 CAAC 的 FDR 数据...23 2.3.1 FDR 图表...25 附录 A. CVR 质量评级量表...28 附录 B. 经验证参数...29

A. ACCIDENT

A. 事故

Location: Wuzhou, China Date: March 21, 2022 Time: 0630 UTC Airplane: Boeing 737-800, China Eastern Airlines, Registration B-1791

地点：Wuzhou, China 日期：March 21, 2022 时间：0630 UTC 飞机：Boeing 737-800，China Eastern Airlines，注册号 B-1791

B. COCKPIT VOICE AND FLIGHT DATA RECORDER DATA RECOVERY GROUP

B. 驾驶舱语音和飞行数据记录器数据恢复小组

Report Author Charles Cates Mechanical Engineer/Recorder Specialist National Transportation Safety Board (NTSB)

报告作者 Charles Cates 机械工程师/记录器专家 National Transportation Safety Board (NTSB)

Group Member Xiangdong Wan Team Leader, Chief Pilot of CAAC Civil Aviation Administration of China (CAAC)

小组成员 Xiangdong Wan 组长，CAAC 主任飞行员 Civil Aviation Administration of China (CAAC)（中国民用航空局）

Group Member Hang Lin Investigator In Charge (IIC) CAAC

小组成员 Hang Lin 主任调查员（IIC） CAAC

Group Member Yu Zhang Investigator CAAC

小组成员 Yu Zhang 调查员 CAAC

Group Member Liling Yu Lab Engineer CAAC

小组成员 Liling Yu 实验室工程师 CAAC

Group Member Xin Miao Lab Engineer CAAC

小组成员 Xin Miao 实验室工程师 CAAC

Group Member Chun Wang Lab Engineer CAAC

小组成员 Chun Wang 实验室工程师 CAAC

Subject Matter Expert Joseph Gregor, Ph.D. Electrical Engineer/Recorder Specialist NTSB

主题专家 Joseph Gregor, Ph.D. 电气工程师/记录器专家 NTSB

Subject Matter Expert R. Greg Smith Branch Chief, Vehicle Recorders Division (Blue) NTSB

主题专家 R. Greg Smith Vehicle Recorders Division (Blue) 分部主任 NTSB

Specialist W. Deven Chen Electrical Engineer/Recorder Specialist NTSB

专家 W. Deven Chen 电气工程师/记录器专家 NTSB

C. DETAILS OF THE INVESTIGATION

C. 调查详情

A data recovery group was convened on March 28, 2022, consisting of representatives from the Civil Aviation Administration of China (CAAC) and the National Transportation Safety Board (NTSB). The NTSB Vehicle Recorder Division received the memory modules from a Cockpit Voice Recorder (CVR) and a Flight Data Recorder (FDR). Work on the memory devices was performed by NTSB personnel and overseen and recorded by CAAC personnel. Memory modules from the following recorders were provided to the NTSB:

一个数据恢复小组于 March 28, 2022 召集成立，由 Civil Aviation Administration of China (CAAC) 和 National Transportation Safety Board (NTSB) 的代表组成。NTSB Vehicle Recorder Division 收到了来自一台 Cockpit Voice Recorder (CVR) 和一台 Flight Data Recorder (FDR) 的存储模块。对存储设备的工作由 NTSB 人员执行，并由 CAAC 人员监督和记录。以下记录器的存储模块被提供给 NTSB：

Recorder Manufacturer/Model: Honeywell HFR5-V CVR

记录器制造商/型号：Honeywell HFR5-V CVR

Part Number: 980-6032-001

部件号：980-6032-001

Recorder Serial Number: CVR-04014

记录器序列号：CVR-04014

Recorder Manufacturer/Model: Honeywell HFR5-D FDR

记录器制造商/型号：Honeywell HFR5-D FDR

Part Number: 980-4750-009

部件号：980-4750-009

Recorder Serial Number: FDR-02952

记录器序列号：FDR-02952

D. DATA RECOVERY

D. 数据恢复

Data recovery from damaged flight data recorders is a methodical process of evaluation, repair, and readout. The highest priority is placed on recovering stored data using a method that is least likely to result in lost or corrupted data. Extensively damaged hardware is a particular challenge due to the possibility of shorted or otherwise damaged board components leading to data loss when power is applied. Careful handling and thorough inspections of components prior to readout attempts can find and catalog damage so that a methodical repair and recovery plan can be pursued. The use of tools specifically designed for recovery of damaged hardware is essential to preserving the integrity of the recorded data. This includes hardware such as cables and readout chassis as well as recovery software packages.

从受损的飞行数据记录器中恢复数据，是一个有条理的评估、修理和读出过程。最高优先级是采用最不可能导致数据丢失或损坏的方法来恢复已存储的数据。严重受损的硬件是一项特别的挑战，因为在通电时，短路或其他受损的电路板部件可能导致数据丢失。在尝试读出之前，对部件进行谨慎处理和彻底检查，可以发现并登记损坏情况，从而能够执行有条理的修理和恢复计划。使用专门为受损硬件恢复而设计的工具，对于保持所记录数据的完整性至关重要。这包括电缆和读出机箱等硬件，以及恢复软件包。

1.0 HFR5-V CVR Description

1.0 HFR5-V CVR 说明

The Honeywell HFR5-V Cockpit Voice Recorder records four channels of high-quality audio information from the Captain's audio panel, the First Officer's audio panel, the Cockpit Observer's audio panel, and the Cockpit Area Microphone (CAM). The digital recording is stored on a solid-state memory module. The channels sourced from the audio panels record for two hours, and the CAM channel records for three hours. The HFR5-V is designed to meet the crash-survivability requirements laid out in EUROCAE standard ED-112A.

Honeywell HFR5-V Cockpit Voice Recorder 记录来自机长音频面板、副驾驶音频面板、驾驶舱观察员音频面板以及 Cockpit Area Microphone (CAM) 的四个通道高质量音频信息。数字记录存储在固态存储模块上。来自音频面板的通道记录 2 小时，CAM 通道记录 3 小时。HFR5-V 设计为满足 EUROCAE 标准 ED-112A 中规定的耐撞性要求。

1.1 HFR5-V CVR Damage

1.1 HFR5-V CVR 损坏

The CVR Crash Survivable Memory Unit (CSMU) was reported to have been recovered from the aircraft wreckage on March 23, 2022. The recorder was heavily damaged by impact forces. The CSMU was opened and the memory module was removed at the CAAC's facilities in Beijing. The red Room Temperature Vulcanizing (RTV) sealant protecting the memory board Circuit Card Assembly (CCA) was removed and several attempts to download the memory contents of the device were made by the CAAC using a surrogate recorder chassis. It was reported that the software annunciated numerous errors during the decompression process, and none of the attempts resulted in intelligible .wav audio files.

据报告，CVR Crash Survivable Memory Unit (CSMU) 于 March 23, 2022 从飞机残骸中被找到。该记录器因冲击力而严重受损。CSMU 在 CAAC 位于 Beijing 的设施内被打开，并取出了存储模块。保护存储板 Circuit Card Assembly (CCA) 的红色 Room Temperature Vulcanizing (RTV) 密封剂被移除，CAAC 使用替代记录器机箱多次尝试下载该设备的存储内容。据报告，在解压缩过程中软件提示了大量错误，且所有尝试均未生成可听懂的 .wav 音频文件。

1.2 HFR5-V CVR Recovery

1.2 HFR5-V CVR 恢复

On March 28, 2022, the NTSB was given a CVR download file in .dlu format from one of the download attempts of the CVR prior to arrival at the NTSB. The file was decompressed in the NTSB CVR lab using Honeywell's recorder recovery software Playback32. Numerous errors were presented during the decompression process. At the end of the decompression, four .wav audio files were generated. The lengths and sample rates of the four files were as expected; however, when played back the files were unintelligible, with stuttering and echoing artifacts and digital noise throughout, and the audio data were unusable.

March 28, 2022，NTSB 收到一个 .dlu 格式的 CVR 下载文件，该文件来自 CVR 在抵达 NTSB 之前的一次下载尝试。该文件在 NTSB CVR 实验室使用 Honeywell 的记录器恢复软件 Playback32 进行解压缩。解压缩过程中出现大量错误。解压缩结束时，生成了 4 个 .wav 音频文件。4 个文件的长度和采样率均符合预期；然而，回放时这些文件无法听懂，整个过程中存在卡顿和回声伪影以及数字噪声，音频数据无法使用。

The CVR memory module was presented to the NTSB for microscopic inspection. Figure 1 shows the connector side of the board and Figure 2 shows the opposite side. Both sides have FLASH memory chips containing data. Preliminary findings showed the board to have substantial damage to the connector area of the device. Numerous solder pads attaching the connector to the board and providing the electrical signal path to the data chips were found bent or completely dislodged from the board. (See Attachment 1, CMM Recovery Report, for additional details and specifics of connector damage). The damaged pins corresponded to chip data lines and address lines. This damage would be significant to the download and is consistent with the audio artifacts present in the data. Damage to chip data lines would manifest as quantization error-type digital noise and damage to chip address lines would manifest as stuttering, echoing, and repeating-type noises.

CVR 存储模块被提交给 NTSB 进行显微检查。Figure 1 显示电路板的连接器一侧，Figure 2 显示相对一侧。两侧均有包含数据的 FLASH 存储芯片。初步发现显示，该电路板在设备的连接器区域存在严重损坏。发现将连接器固定到电路板并为数据信号芯片提供电气信号路径的许多焊盘已弯曲或完全从电路板上脱落。（有关连接器损坏的更多细节和具体情况，见 Attachment 1, CMM Recovery Report）。受损引脚对应于芯片数据线和地址线。该损坏会对下载产生重大影响，并与数据中存在的音频伪影一致。芯片数据线损坏会表现为量化误差类型的数字噪声，芯片地址线损坏会表现为卡顿、回声和重复类型的噪声。

Further inspections also revealed that the plastic portion of the connector itself was deformed subtly out of its normal state. The long edge along the board end was

进一步检查还显示，连接器本身的塑料部分已轻微变形，偏离其正常状态。沿电路板端部的长边已

Figure 1. CVR memory module board, connector side.

Figure 1. CVR 存储模块电路板，连接器侧。

Figure 2. CVR memory board, side opposite connector.

Figure 2. CVR 存储板，与连接器相对的一侧。

raised above its normal position, and the pins along that edge were recessed inside the connector housing. In this state it would have been difficult for the male pins on the recovery cable to make a reliable connection with the board. The long edge facing the memory chips was crushed down below its normal position with the pins projecting out notably from the connector housing. It was also found that the ground plane between the two sides of the connector was damaged underneath the connector housing. Several sections of the ground plane were severed or lifted from the board. Figure 3 shows the connector, as received.

被抬高到其正常位置以上，并且沿该边缘的插针凹入连接器外壳内部。在这种状态下，恢复电缆上的公插针将很难与电路板形成可靠连接。面向存储芯片的长边被压碎至低于其正常位置，插针明显从连接器外壳中伸出。还发现连接器两侧之间的接地平面在连接器外壳下方受损。接地平面的若干部分被切断或从电路板上翘起。Figure 3 显示了收到时的连接器状态。

 Figure 3. Detail of CVR memory module connector, as received.

 Figure 3. 收到时的 CVR 存储模块连接器细节。

There was one additional anomaly noted that was determined to have been present at the time of manufacture of the board. A capacitor was lifted from the board and not in contact with the board pads, and it had been sealed over with conformal coating in this fashion. This means that the capacitor had never been a part of the circuit it was designed to be in. Review of board schematics showed that this was a power conditioning capacitor, serving to filter frequency fluctuations that may be caused by a dirty power supply. Because it had not been detected in board pass-off testing or continued airworthiness checks of the CVR system performed by the airline, it is unlikely that it impacted the performance of the CVR.

还注意到另一个异常，经确定该异常在电路板制造时即已存在。一个电容器从电路板上翘起，未与电路板焊盘接触，并且以这种状态被保形涂层覆盖密封。这意味着该电容器从未成为其设计所在电路的一部分。对电路板原理图的审查显示，这是一个电源调节电容器，用于滤除可能由不洁净电源造成的频率波动。由于该问题在电路板交付验收测试中未被发现，也未在航空公司对 CVR 系统进行的持续适航检查中被发现，因此它不太可能影响 CVR 的性能。

To finalize the preliminary inspections of the board, the remaining RTV sealant on the board was carefully removed from the connector area. All the pins from the connector that were accessible for inspection were physically tested for integrity. (Some pins were not accessible because they were routed under the board.) Numerous loose or separated pins were catalogued. Figure 4 shows the connector prepped for inspection with remaining RTV sealant removed.

为完成对该电路板的初步检查，仔细清除了电路板上连接器区域剩余的 RTV 密封胶。对连接器上所有可检查的插针进行了完整性物理测试。（有些插针不可接近，因为其走线位于电路板下方。）记录了许多松动或分离的插针。Figure 4 显示了清除剩余 RTV 密封胶后、准备接受检查的连接器。

A damage recovery work plan was developed to attempt to repair the damage discovered during preliminary inspections as safely as possible before powering the board and attempting another download of the memory. The circuit board Gerber files were reviewed to determine the location and routing of all the significant vias and memory traces. A continuity test was developed to check for hidden electrical damage between the connector pins and the flash memory chips. The test was confirmed using a known good surrogate memory module.

制定了一项损伤恢复工作计划，以便在给电路板上电并尝试再次下载存储器之前，尽可能安全地修复初步检查中发现的损伤。审查了电路板 Gerber 文件，以确定所有重要过孔和存储器走线的位置及布线。制定了一项连续性测试，用于检查连接器插针与闪存芯片之间是否存在隐藏的电气损伤。该测试使用一个已知良好的替代存储模块进行了确认。

 Figure 4. Connector detail with RTV sealant removed.

 Figure 4. 已清除 RTV 密封胶的连接器细节。

The loose and separated pins found in the inspections were repaired using a microscope and a special cyanoacrylate glue. To help make the best connection possible with the damaged board connector, an HFR-5 recovery flex cable was modified to aid in positive connections between the cable and the board. Many of the pins on the cable were bent slightly away from the plastic of the connector, causing them to interface better with the pins that were recessed into the damaged board connector.

检查中发现的松动和分离插针使用显微镜和一种专用氰基丙烯酸酯胶进行了修复。为帮助与受损电路板连接器形成尽可能良好的连接，对一根 HFR-5 恢复柔性电缆进行了改装，以帮助在电缆与电路板之间形成可靠连接。电缆上的许多插针被略微弯离连接器塑料件，使其能够更好地与凹入受损电路板连接器内的插针对接。

With these repairs complete, the accident CPM was interfaced with the modified recovery flex cable and installed in the NTSB's HFR5-V golden chassis. A download was attempted using Honeywell Playback32 software, which failed to generate a usable .dlu file.

完成这些修复后，将事故 CPM 与改装后的恢复柔性电缆连接，并安装到 NTSB 的 HFR5-V golden chassis（基准机箱）中。使用 Honeywell Playback32 软件尝试下载，但未能生成可用的 .dlu 文件。

A subsequent download was performed without unplugging or reseating the connector from the golden chassis using an engineering tool called DLDR. DLDR provides a direct read of the flash memory chip contents and writes each chip image into binary chip image files. The chip images can then be recombined in the proper sequence into the full memory contents and written into a .dlu file using another engineering tool called CHIPS. This download generated 14 chip image files (one for each flash memory chip on the board, as expected), which were combined into a .dlu file. Playback32 was able to generate 4 .wav files from the images; however, it generated a large volume of errors during the decompression process. A second DLDR download was completed to check for consistency between the two downloads. The downloads had minor single-byte differences spaced throughout the file, which were never explained.

随后在不从 golden chassis 拔下或重新插接连接器的情况下，使用名为 DLDR 的工程工具执行了一次下载。DLDR 可直接读取闪存芯片内容，并将每个芯片镜像写入二进制芯片镜像文件。随后可使用另一种名为 CHIPS 的工程工具，将这些芯片镜像按正确顺序重新组合成完整存储器内容，并写入 .dlu 文件。此次下载生成了 14 个芯片镜像文件（如预期，对应电路板上的每个闪存芯片各一个），这些文件被合并成一个 .dlu 文件。Playback32 能够从这些镜像生成 4 个 .wav 文件；但是，在解压缩过程中产生了大量错误。完成了第二次 DLDR 下载，以检查两次下载之间的一致性。两次下载在整个文件中分散存在轻微的单字节差异，其原因一直未得到解释。

The .wav files included three audio panel channels sampled at 8 kHz of two hours in length each and one CAM channel sampled at 16 kHz of three hours in length. The files were loaded into a CVR listening room for audition. Because of the method of recombining chip image files, the .wav files were not in sequential order (i.e., with the oldest data at the beginning running sequentially through the most recent data at the end). Instead, the data was in the order in which it was written across the chips, and the most recent data (the accident sequence) was near the middle of the files, followed by the oldest data.

这些 .wav 文件包括三个音频面板通道，每个通道以 8 kHz 采样、长度为两小时，以及一个 CAM 通道，以 16 kHz 采样、长度为三小时。这些文件被加载到 CVR 监听室进行试听。由于重新组合芯片镜像文件的方法，这些 .wav 文件并非按顺序排列（即最旧数据在开头，并依次延续到结尾的最新数据）。相反，数据是按照其写入各芯片的顺序排列，最新数据（事故序列）位于文件的中间附近，其后是最旧数据。

The audition showed that all four channels contained intelligible audio. The audio panel channels were of Fair quality on the NTSB quality rating scale¹, and the CAM channel was of Poor quality. There was digital noise and distortion present on all channels, and there were portions of repeated audio noted during the accident sequence. The digital artifacts in the audio data were consistent with damage between the recovery cable and the NVM chip data lines, and the repeated audio was consistent with similar damage to the chip address lines.

试听显示，所有四个通道均包含可理解的音频。按 NTSB 质量评级量表¹，音频面板通道质量为 Fair（一般），CAM 通道质量为 Poor（差）。所有通道均存在数字噪声和失真，并且在事故序列期间注意到有部分重复音频。音频数据中的数字伪影与恢复电缆和 NVM 芯片数据线之间的损伤相一致，而重复音频与芯片地址线的类似损伤相一致。

The download results provided additional evidence of damage to the connector area of the board, so further inspections were developed to identify the issues with the pin connections. These included both 2-D and 3-D X-ray imaging to determine if any non-visible damage had occurred to the memory module and additional microscopic inspections.

下载结果为电路板连接器区域受损提供了进一步证据，因此制定了进一步检查，以识别插针连接存在的问题。这些检查包括 2-D 和 3-D X 射线成像，以确定存储模块是否发生了任何不可见损伤，并包括额外的显微检查。

X-ray inspections showed several instances where the pins of the connector had minimal clearance between either board traces or vias; however, the clearance indicated it was unlikely to lead to shorts or other problems with the signal path of the chip data and address lines. No other damage was seen in the additional microscopic inspections.

X 射线检查显示，连接器插针与电路板走线或过孔之间在若干处间隙很小；但是，该间隙表明其不太可能导致短路或芯片数据线和地址线信号路径的其他问题。在额外的显微检查中未发现其他损伤。

The modified recovery ribbon cable was adjusted again with the help of the X-ray images in an attempt to make better connections with the damaged connector. A comprehensive pin check was undertaken to ensure that the pins with minimal clearance were not causing issues with the electrical signal path. Another series of downloads was attempted, with similar results to the previous downloads.

在 X 射线图像的帮助下，再次调整了改装后的恢复带状电缆，试图与受损连接器形成更好的连接。进行了全面的插针检查，以确保间隙很小的插针未对电气信号路径造成问题。又尝试了一系列下载，结果与之前的下载类似。

¹ Appendix A comprises the CVR Quality Rating Scale.

¹ Appendix A 包含 CVR Quality Rating Scale。

With the downloads only providing poor to fair results and the audio being consistent with damage to chip data and address lines even after attempts to correct the issue, the team determined that the damaged board connector should be removed and possibly replaced. An NTSB surrogate board was identified to be used as a test article to determine the safest process to remove the plastic housing of the connector shell while leaving the pins soldered in place on the board.

由于下载结果仅为差到一般，且即使在尝试纠正问题后音频仍与芯片数据线和地址线损伤相一致，团队确定应拆除并可能更换受损的电路板连接器。确定使用一块 NTSB 替代电路板作为试验件，以确定在使插针仍焊接在电路板上的情况下，拆除连接器外壳塑料壳体的最安全工艺。

The CVR connector housing was fully removed, which revealed additional damage to the pins that was hidden underneath the plastic. Pins were found pushed down underneath the plastic and numerous additional pads and traces were found to be loose or completely removed from the board. It is likely that the only reason some pins were making contact was that they were pinned down to the board by the connector housing itself. The central ground planes were found to be loose at every attachment point except for one. The memory module with the connector shell removed is shown in Figure 5. Some of the deformations to the pins are shown in Figure 6.

CVR 连接器外壳被完全拆除，这揭示了隐藏在塑料下方的插针额外损伤。发现插针被压到塑料下方，并发现许多额外的焊盘和走线松动或已从电路板上完全脱离。很可能有些插针能够接触的唯一原因，是其被连接器外壳本身压在电路板上。发现中央接地平面除一个连接点外，在所有连接点处均已松动。拆除连接器外壳后的存储模块如 Figure 5 所示。插针的一些变形如 Figure 6 所示。

 Figure 5. Detail view of connector pins with shell removed.

 Figure 5. 拆除外壳后的连接器插针详细视图。

 Figure 6. Detailed view of deformed pins.

 Figure 6. 变形插针的详细视图。

The pins, now without the plastic connector shell, were again stabilized and loose areas were reworked with cyanoacrylate glue and allowed to cure. With the pins completely stabilized, work was undertaken to carefully re-shape them into their proper positions. The pins on the surrogate board were used as a template for this activity. Re-shaping the pins into the best approximation of their original shapes provided the best chance of securely mating to the recovery cable once the plastic housing was replaced. With the pins re-shaped, it also became clear that the odd-numbered pins were critically close to many of the uninsulated vias and traces, which was previously hidden by the connector shell. To solve this issue, a thin insulation strip was fabricated out of Kapton tape and placed between the pins and the board.

这些插针此时已没有塑料连接器外壳，再次对其进行稳定处理，并用氰基丙烯酸酯胶对松动区域进行返工并使其固化。在插针完全稳定后，开展工作以小心地将其重新整形成适当位置。使用替代电路板上的插针作为该工作的模板。将插针重新整形成尽可能接近其原始形状，为塑料外壳更换后与恢复电缆牢固配合提供了最佳机会。插针重新整形后，还可以清楚看到奇数编号插针与许多未绝缘过孔和走线极其接近，而这此前被连接器外壳所遮挡。为解决这一问题，使用 Kapton 胶带制作了一条薄绝缘条，并将其置于插针与电路板之间。

A new connector shell was carefully installed over the reworked pins. This work was performed under the microscope due to the delicate nature of the pins and the need to have them align with the individual grooves in the connector. The modified flex cable was mated to the board and secured with ESD tape. The board and flex cable assembly was then attached to the NTSB HFR5-V golden chassis and downloaded using both Playback32 and DLDR.

小心地将一个新的连接器外壳安装在返工后的插针上方。由于插针十分脆弱且需要与连接器中的各个凹槽对齐，该工作在显微镜下进行。将改装后的柔性电缆与电路板对接，并用 ESD 胶带固定。随后将电路板和柔性电缆组件连接到 NTSB HFR5-V golden chassis，并使用 Playback32 和 DLDR 进行下载。

The Playback32 download proceeded normally with none of the errors seen in previous download attempts. Likewise, the decompression process went smoothly with a small number of errors during the decompression of the wideband (CAM) channel, but none in the narrowband channels, or in the sheer quantity that was seen in previous downloads. Playback32 generated 4 .wav files with the bit depth and sampling rate characteristics equivalent to previous downloads, as expected.

Playback32 下载正常进行，未出现之前下载尝试中所见的错误。同样，解压缩过程也顺利进行，在宽带（CAM）通道解压缩期间出现少量错误，但窄带通道未出现错误，也未出现之前下载中所见的大量错误。如预期，Playback32 生成了 4 个 .wav 文件，其位深和采样率特性与之前下载相当。

The .wav files were loaded into a CVR listening room for audition. The files were in sequential order from the oldest data to the most recent (accident) data. The accident was captured, and audio quality was Excellent per the NTSB audio quality rating standards on all channels.

这些 .wav 文件被加载到 CVR 监听室进行试听。文件按从最旧数据到最新（事故）数据的顺序排列。事故被记录下来，并且按照 NTSB 音频质量评级标准，所有通道的音频质量均为 Excellent（优秀）。

Further analysis of the individual tracks revealed that there was a time drift between the wideband CAM channel and the narrowband crew audio channels. The drift was not linear, and the CAM and crew channels drifted from each other throughout the full two hours of the crew channel recordings. When a common audio anchor point was used to synchronize the CAM and crew channels, it could be determined that the crew channels were occasionally missing small portions of data, likely corresponding with individual packets of digital audio data being dropped in the recovery.

对各单独音轨的进一步分析显示，宽带 CAM 通道与窄带机组音频通道之间存在时间漂移。该漂移并非线性，并且在机组通道录音的完整两小时内，CAM 与机组通道彼此发生漂移。当使用一个共同音频锚点来同步 CAM 与机组通道时，可以确定机组通道偶尔缺失小段数据，很可能对应于恢复过程中个别数字音频数据包丢失。

The DLDR download was also decompressed and .wav files were generated with the files from the DLDR-generated chip images. The .wav files were directly compared with the .wav files generated from the Playback32 download. It was determined that the Playback32 files provided the most accurate time history for the event. The narrowband channels exhibited minor data loss for very short durations occurring at random intervals throughout the recording. These artifacts may have been related to the as-manufactured bypass capacitor damage identified on the CVR board during initial microscopic inspection, but this could not be confirmed because there was no exemplar download from before the accident for comparison.

DLDR 下载也进行了扩展，并使用 DLDR 生成的芯片镜像文件生成了 .wav 文件。将这些 .wav 文件与从 Playback32 下载生成的 .wav 文件进行了直接比较。经确定，Playback32 文件为该事件提供了最准确的时间历程。窄带通道表现出轻微数据丢失，持续时间很短，并在整个录音中以随机间隔发生。这些伪影可能与最初显微检查期间在 CVR 电路板上识别出的制造状态旁路电容损伤有关，但由于没有事故前的示例下载可供比较，无法确认这一点。

With excellent quality recordings and an accurate time history on the CAM channel, it was determined that an acceptable download was recovered, and no additional download attempts or more invasive recovery techniques were required.

鉴于获得了优秀质量的录音，且 CAM 通道具有准确的时间历程，确定已恢复了可接受的下载数据，不需要进行额外下载尝试或更具侵入性的恢复技术。

## 1.3 Audio Recording Description

## 1.3 音频记录说明

Each channel's audio quality and duration from the final download files is indicated in Table 1.

最终下载文件中每个通道的音频质量和时长见 Table 1。

Table 1: Audio Quality | Channel Number | Content/Source | Quality | Duration | | --- | --- | --- | --- | | 1 | Cockpit Observer Audio Panel | Excellent | ~120 min | | 2 | First Officer Audio Panel | Excellent | ~120 min | | 3 | Captain Audio Panel | Excellent | ~120 min | | 4 | Cockpit Area Microphone (CAM) | Excellent | ~180 min |

Table 1: 音频质量 | 通道编号 | 内容/来源 | 质量 | 时长 | | --- | --- | --- | --- | | 1 | 驾驶舱观察员音频面板 | Excellent（优秀） | ~120 min | | 2 | 副驾驶音频面板 | Excellent（优秀） | ~120 min | | 3 | 机长音频面板 | Excellent（优秀） | ~120 min | | 4 | Cockpit Area Microphone (CAM)（驾驶舱区域麦克风） | Excellent（优秀） | ~180 min |

## 1.4 CVR Data Provided to CAAC

## 1.4 提供给 CAAC 的 CVR 数据

CVR data provided to the CAAC delegation included the following:

提供给 CAAC 代表团的 CVR 数据包括以下内容：

- Raw download files from all download attempts made, both with Playback32 and DLDR - Raw chip image files from all DLDR download attempts - All .dlu files generated from both Playback32 and DLDR/CHIPS, including those that did not generate valid .wav files - All .wav files generated from the downloads. - All .wav files that were timeline-corrected and all crew audio channels that were upsampled from their default 8 kHz sample rate to match the CAM channel sample rate at 16 kHz. - All session files generated by the NTSB's audio analysis tool, known as RAPT-R .mdf files. - .wav files that included noise filtering and audio enhancements to help bring out crucial portions of the audio. - Time-stretched .wav files to match the timing of the crew channels and the CAM channel as best as possible given the packet dropouts discussed in the text above. - All photographs, scans, and microscopic images taken of the CVR memory boards throughout the recovery process.

- 使用 Playback32 和 DLDR 进行的所有下载尝试的原始下载文件 - 所有 DLDR 下载尝试的原始芯片镜像文件 - 由 Playback32 和 DLDR/CHIPS 生成的所有 .dlu 文件，包括未生成有效 .wav 文件的文件 - 下载生成的所有 .wav 文件。 - 所有经过时间线校正的 .wav 文件，以及所有从默认 8 kHz 采样率上采样到与 CAM 通道 16 kHz 采样率匹配的机组音频通道。 - 由 NTSB 的音频分析工具生成的所有会话文件，即 RAPT-R .mdf 文件。 - 包含降噪和音频增强处理、用于帮助突出音频关键部分的 .wav 文件。 - 时间拉伸后的 .wav 文件，在考虑上文所述数据包丢失的情况下，尽可能匹配机组通道与 CAM 通道的时间。 - 在整个恢复过程中拍摄或扫描的 CVR 存储电路板的所有照片、扫描件和显微图像。

The NTSB did not retain any of the above files provided to the CAAC delegation other than the photographs, scans, and microscopic images. No CVR audio files or other raw or intermediate download files that could be used to generate audio files were maintained by the NTSB.

除照片、扫描件和显微图像外，NTSB 未保留上述提供给 CAAC 代表团的任何文件。NTSB 未保留任何 CVR 音频文件，也未保留可用于生成音频文件的其他原始或中间下载文件。

## 2.0 HFR5-D FDR Description

## 2.0 HFR5-D FDR 说明

The Honeywell HFR5-D Flight Data Recorder records airplane flight information in a digital format using solid-state flash memory as the recording medium. The HFR5-D can receive data in the ARINC 573/717/747 configurations and can record a minimum of 25 hours of flight data. It is configured to record 512 12-bit words of digital information every second. Each grouping of 512 words (each second) is called a subframe. Each subframe has a unique 12-bit synchronization (sync) word identifying it as either subframe 1, 2, 3, or 4. The sync word is the first word in each subframe. The data stream is "in sync" when successive sync words appear at proper 512-word intervals. Each data parameter (e.g., altitude, heading, airspeed) has a specifically assigned word number within the subframe. The HFR5-D is designed to meet the crash-survivability requirements of EUROCAE standard ED-112A.

Honeywell HFR5-D Flight Data Recorder（飞行数据记录器）使用固态闪存作为记录介质，以数字格式记录飞机飞行信息。HFR5-D 可接收 ARINC 573/717/747 配置的数据，并可记录至少 25 小时的飞行数据。其配置为每秒记录 512 个 12-bit 字的数字信息。每组 512 个字（每秒）称为一个子帧。每个子帧都有一个唯一的 12-bit 同步（sync）字，用于将其识别为子帧 1、2、3 或 4。同步字是每个子帧中的第一个字。当连续的同步字以正确的 512 字间隔出现时，数据流处于“in sync”状态。每个数据参数（例如高度、航向、空速）在子帧内都有专门指定的字编号。HFR5-D 设计为满足 EUROCAE standard ED-112A 的耐坠毁要求。

## 2.1 HFR5-D FDR Damage

## 2.1 HFR5-D FDR 损伤

The FDR crash survivable memory unit (CSMU) was reported to have been recovered from the aircraft wreckage on March 27, 2022. The recorder was heavily damaged by impact forces. The CSMU was opened and the memory module was

据报告，FDR crash survivable memory unit (CSMU)（耐坠毁存储单元）于 March 27, 2022 从飞机残骸中被找到。该记录器因撞击力严重受损。CSMU 被打开，存储模块被

removed at the CAAC's facilities in Beijing. It was reported that one attempt to download the memory contents was made by the CAAC using a surrogate FDR chassis, which was unsuccessful.

在 CAAC（中国民用航空局）位于 Beijing（北京）的设施中被拆下。据报告，CAAC 曾使用替代 FDR 机箱尝试下载一次存储器内容，但未成功。

## 2.2 HFR5-D FDR Recovery

## 2.2 HFR5-D FDR 恢复

The FDR CCA was presented to the NTSB on April 4, 2022. The red RTV sealant protecting the memory board CCA was left intact, with only the FLASH memory temperature dot exposed for inspection. The temperature dot indicated no significant heat exposure to the board and none of the damage present on the CVR board was noted on the FDR.

FDR CCA 于 2022 年 4 月 4 日提交给 NTSB（美国国家运输安全委员会）。用于保护存储器板 CCA 的红色 RTV 密封胶保持完好，仅露出 FLASH 存储器温度指示点以供检查。该温度指示点表明电路板未受到显著热暴露，且在 FDR 上未发现 CVR 电路板上存在的损伤。

The NTSB golden chassis was reconfigured to be a proper FDR surrogate and the accident memory was connected. A download attempt using Playback32 was attempted, which failed. Several downloads were performed of the individual FLASH memory devices using the DLDR tool. The results varied slightly from one download attempt to the next; however, they generally resulted in files of all 0's for each FLASH chip image.

NTSB 的标准参考机箱被重新配置为合适的 FDR 替代机箱，并连接了事故存储器。使用 Playback32 尝试进行下载，但失败。随后使用 DLDR 工具对各个 FLASH 存储器器件进行了数次下载。每次下载尝试的结果略有不同；但通常每个 FLASH 芯片映像得到的文件均为全 0。

The red RTV was removed from the FDR memory board to allow for an initial visual inspection of the entire module. Figures 7 and 8 show the connector side and opposite side, respectively, of the FDR memory module. The module was inspected under high-power microscopy, revealing that FLASH module number U2 had a crack extending from the top edge of the memory packaging into the center of the chip. The crack penetrated to the surface of the conformal coating over the device. Many of the epoxy balls bonding the chip to the board also showed evidence of cracking and/or separation from the FLASH devices to which they were originally adhered.

从 FDR 存储器板上去除了红色 RTV，以便对整个模块进行初步目视检查。图 7 和图 8 分别显示 FDR 存储器模块的连接器侧和相对侧。该模块在高倍显微镜下进行了检查，发现编号为 U2 的 FLASH 模块有一条裂纹，从存储器封装的上边缘延伸至芯片中心。该裂纹贯穿至器件上方保形涂层的表面。许多将芯片粘接到电路板上的环氧树脂胶球也显示出开裂和/或与其原本粘附的 FLASH 器件分离的迹象。



Figure 7. Connector side of FDR memory module with crack faintly visible in chip U2, marked with a red circle in the photograph.

图 7. FDR 存储器模块的连接器侧，芯片 U2 中可隐约看到裂纹，照片中用红圈标出。



Figure 8. FDR memory module side opposite connector.

图 8. FDR 存储器模块与连接器相对的一侧。

Additional minor cracking was found in the chip packaging areas of many of the epoxy bonds between FLASH modules. This included minor cracking in FLASH module numbers U1 & U2, U5 & U7, and U51 & U52. The majority of the epoxy bonds on both sides of the memory module showed evidence of fractures. Figure 9 shows an example of the cracking seen in the epoxy bond between chips U1 and U2.

在 FLASH 模块之间许多环氧树脂粘接处的芯片封装区域发现了额外的轻微开裂。这包括 FLASH 模块编号 U1 & U2、U5 & U7 以及 U51 & U52 处的轻微开裂。存储器模块两侧的大多数环氧树脂粘接处均显示出断裂迹象。图 9 显示了芯片 U1 和 U2 之间环氧树脂粘接处所见开裂的示例。



Figure 9. Example of cracks seen in epoxy bond between chips U1 and U2.

图 9. 芯片 U1 和 U2 之间环氧树脂粘接处所见裂纹示例。

X-ray inspection of the FDR module was undertaken to evaluate any damage not detectable via optical examination. This damage could include broken bond wires within the device packaging, damage to the frame of the chip, or damage to the chip silicon die. No additional damage was discovered during this inspection.

对 FDR 模块进行了 X 射线检查，以评估任何无法通过光学检查检测到的损伤。此类损伤可能包括器件封装内键合线断裂、芯片框架损坏，或芯片硅裸片损坏。本次检查未发现其他损伤。

Following the discovery of chip-level cracking and the behavior of the downloads using the DLDR chip-level recovery tool, there was concern for the integrity of the data using additional standard download methods. Following discussions among staff, the conclusion was reached that the safest method for recovery of any surviving data would be to remove the individual FLASH memory devices (chips) containing flight data and read the data directly using a FLASH memory chip reader.

在发现芯片级开裂以及使用 DLDR 芯片级恢复工具下载时的表现之后，使用其他标准下载方法所得数据的完整性引起了担忧。经工作人员讨论后，得出结论：恢复任何幸存数据的最安全方法，是拆下包含飞行数据的各个 FLASH 存储器器件（芯片），并使用 FLASH 存储器芯片读取器直接读取数据。

Removal of FLASH memory devices from a circuit board assembly is a technically challenging and somewhat risky procedure. It is possible to damage the chips both physically and electrically during the removal and readout process. As a risk-reduction exercise, the NTSB decided to use a known good surrogate memory module as a test article to develop the safest procedure for removal and readout of the individual FLASH memory modules.

从电路板组件上拆除 FLASH 存储器器件是一项技术上具有挑战性且有一定风险的程序。在拆除和读出过程中，可能会对芯片造成物理和电气损伤。作为降低风险的措施，NTSB 决定使用一个已知良好的替代存储器模块作为试件，以制定拆除和读出各个 FLASH 存储器模块的最安全程序。

A suitable NTSB surrogate test article was identified, and the memory was first downloaded using the DLDR tool. This would permit direct comparison of the chip images prior to removal from the board with the images generated with a FLASH memory reader/programmer. If any differences existed, it would be possible to characterize and remediate differences between the two chip readout methods.

确定了一个合适的 NTSB 替代试件，并首先使用 DLDR 工具下载其存储器。这样可以将从电路板拆除之前的芯片映像与使用 FLASH 存储器读取器/编程器生成的映像进行直接比较。如果存在任何差异，则可以对两种芯片读出方法之间的差异进行表征并加以修正。

NTSB electronics specialists removed all of the conformal coating from the FLASH memory pins of the test article in preparation for de-soldering the chips from the PCB. Then, all the epoxy bonds were manually removed from each of the chips. A first pass at solder removal was made by hand, prior to using an automated tool. This was done to reduce the stress of removal of the chips. It was expected that there would still be residual RTV, conformal coating, and epoxy trapped underneath the chip where it was placed on the PCB.

NTSB 电子专家清除了试件 FLASH 存储器引脚上的全部保形涂层，为从 PCB 上拆焊芯片做准备。随后，手工去除了每个芯片上的所有环氧树脂粘接。先手工进行了第一遍焊料去除，然后再使用自动化工具。这样做是为了降低拆除芯片时的应力。预计在芯片放置于 PCB 的位置下方仍会有残留的 RTV、保形涂层和环氧树脂滞留。

Finally, a recipe was developed for the Finetech FinePlacer microelectronics workstation for chip removal. The recipe consisted of the following:

最后，为 Finetech FinePlacer 微电子工作站制定了用于芯片拆除的工艺配方。该配方包括以下内容：

1. Heating of the entire board to an equilibrium temperature somewhat below the solder melt temperature for the solder used by the manufacturer. 2. Application of heat to the pins of the memory chip through a chip-specific nozzle that ramped the temperature at a safe rate to reach solder-melt temperatures. 3. Application of suction to lift the chip free of the PCB once the solder was fully melted.

1. 将整块电路板加热至平衡温度，该温度略低于制造商所用焊料的焊料熔化温度。 2. 通过芯片专用喷嘴向存储器芯片引脚施加热量，使温度以安全速率升高至焊料熔化温度。 3. 在焊料完全熔化后施加吸力，将芯片从 PCB 上提起并分离。

On the HFR5 memory board there are a total of 14 FLASH memory chips, 7 on each side of the board. However, in the FDR application (HFR5-D) only 6 of the chips (chips defined as U1, U2, U3, U4, U5, and U6) are used to store flight data. For the first test of removal from the test article, chip U7 was removed. The first try of the Finetech removal recipe worked well and the chip was removed without difficulty.

HFR5 存储器板上共有 14 个 FLASH 存储器芯片，电路板每侧 7 个。然而，在 FDR 应用（HFR5-D）中，仅 6 个芯片（定义为 U1、U2、U3、U4、U5 和 U6 的芯片）用于存储飞行数据。对试件进行第一次拆除测试时，拆下了芯片 U7。首次尝试 Finetech 拆除配方效果良好，芯片顺利拆除。

Following the success of the removal of chip U7, each of the six chips from the test article containing flight data was removed individually in turn with the same recipe. In some cases, additional force was needed to remove the chips from the board because of residual sealant trapped underneath. As each chip was removed, it was cleaned and any remaining residual solder was removed. The chip was then placed into the proper socket for a Xeltek SP-6100 FLASH memory reader/programmer for download via computer.

在成功拆除芯片 U7 后，试件中每一个包含飞行数据的六个芯片均依次使用相同配方单独拆除。在某些情况下，由于下方滞留的残余密封胶，需要施加额外的力才能将芯片从电路板上拆下。每个芯片拆下后均进行清洁，并去除任何剩余的残余焊料。随后将芯片放入 Xeltek SP-6100 FLASH 存储器读取器/编程器的相应插座中，通过计算机进行下载。

The chip images generated by the chip reader were compared to the original chip images generated by the DLDR tool and found to be byte-swapped identical copies of each other. This is because the DLDR tool was designed to create binary chip images in the Big Endian (Intel) format and the chip reader generated images in the Little Endian (Motorola) format. This was not an issue because the Honeywell CHIPS utility is able to read chip images in either format, as selected by the user.

将芯片读取器生成的芯片映像与 DLDR 工具生成的原始芯片映像进行比较，发现二者为字节交换后相同的副本。这是因为 DLDR 工具被设计为以 Big Endian (Intel) 格式创建二进制芯片映像，而芯片读取器以 Little Endian (Motorola) 格式生成映像。这并不是问题，因为 Honeywell CHIPS 实用程序能够按用户选择读取任一格式的芯片映像。

As a final test, DLU files were generated using both the DLDR chip images and the memory reader/programmer chip images. The two DLU files were found to be functionally identical, thus validating the full process for chip-off downloads and data recovery to be used on the accident memory board.

作为最终测试，分别使用 DLDR 芯片映像和存储器读取器/编程器芯片映像生成了 DLU 文件。发现两个 DLU 文件在功能上相同，从而验证了将用于事故存储器板的脱板芯片下载和数据恢复完整流程。

Permission was granted by the CAAC to begin recovery of data from the accident module using the same process developed with the test article. The same process of conformal coating stripping and epoxy removal was used to prepare the chips for removal from the board. Excess solder was again removed by hand before being placed in the FinePlacer tool. The chip removal process with the Finetech FinePlacer went smoothly for all six of the chips of interest.

CAAC 授权开始使用与试件所开发流程相同的流程，从事故模块中恢复数据。采用相同的保形涂层剥除和环氧树脂去除流程，为将芯片从电路板上拆除做准备。在放入 FinePlacer 工具之前，再次手工去除了多余焊料。使用 Finetech FinePlacer 对全部六个目标芯片进行的芯片拆除过程进展顺利。

An inspection of chip U2 (the chip with the visible crack in the packaging) identified a network of cracks on the bottom of the chip in addition to the previously observed crack. The other chips did not reveal additional damage when they were removed from the board. Figure 10 shows the bottom of chip U2 following removal from the board.

对芯片 U2（封装中有可见裂纹的芯片）的检查发现，除先前观察到的裂纹外，芯片底部还存在裂纹网络。其他芯片从电路板上拆下时未发现额外损伤。图 10 显示了从电路板拆除后的芯片 U2 底部。



Figure 10. Bottom side of chip U2 showing cracks throughout chip packaging.

图 10. 芯片 U2 底面，显示整个芯片封装内的裂纹。

The chips were read in the Xeltek memory reader and the resulting binary images for chips U1, U3, U4, U5, and U6 were found to contain data. Chip U2 could not be read and the Xeltek reported numerous pin-check errors and messages indicating that many of the pins were shorted. This was a sign of significant internal damage to the silicon die within the chip itself.

这些芯片在 Xeltek 存储器读取器中读取，发现芯片 U1、U3、U4、U5 和 U6 生成的二进制映像包含数据。芯片 U2 无法读取，Xeltek 报告了大量引脚检查错误，并显示许多引脚短路的消息。这表明芯片自身内部的硅裸片存在严重内部损伤。

Chip U2 was X-rayed and CT scanned, and chip U4 was also X-rayed as an exemplar. The results of the CT scan were noisy and revealed no new information. Some of the cracks identified visually on the bottom side of the packaging were apparent in the 2-D X-ray images. These cracks passed through the region of the chip housing the silicon die containing all the internal circuitry. Several other of the undersurface cracks were not visible in the X-ray because they appeared to follow the outline of the X-shaped copper die pad upon which the silicon die was bonded during manufacture of the chip. This extreme damage rendered any data from U2 unrecoverable.

对芯片 U2 进行了 X 射线检查和 CT 扫描，并对芯片 U4 也进行了 X 射线检查作为参照样本。CT 扫描结果噪声较大，未揭示新的信息。封装底面上通过目视识别出的一些裂纹在 2-D X 射线图像中可见。这些裂纹穿过芯片中容纳硅裸片的区域，该硅裸片包含全部内部电路。还有若干底面裂纹在 X 射线图像中不可见，因为它们似乎沿着 X 形铜裸片焊盘的轮廓延伸，而硅裸片在芯片制造过程中键合在该焊盘上。这种极端损伤导致 U2 中的任何数据均无法恢复。

A dummy file of all hexadecimal 'FF' (binary '1') was created and used in place of the image from U2 in the Honeywell CHIPS utility to create a FDR DLU file. The resultant DLU file was then imported into the NTSB CIDER data analysis application. The data was framed out with information provided by the aircraft manufacturer.

创建了一个全为十六进制“FF”（二进制“1”）的虚拟文件，并在 Honeywell CHIPS utility（Honeywell CHIPS 工具）中用其替代来自 U2 的映像，以创建 FDR DLU 文件。随后将生成的 DLU 文件导入 NTSB CIDER data analysis application（NTSB CIDER 数据分析应用程序）。数据依据 aircraft manufacturer（飞机制造商）提供的信息进行了分帧解析。

When framed out, the accident was found in the middle of the data stream, because the pointer information was lost when the DLU was created in CHIPS without chip U2. The converted data was manually re-ordered to place the accident flight at the end of the recording.

分帧解析后发现，事故位于数据流的中间，因为在 CHIPS 中未使用 U2 芯片创建 DLU 时，指针信息丢失。转换后的数据被手动重新排序，以将事故飞行置于记录的末尾。

Because of the missing chip image there were regular data dropouts throughout the duration of the data file. This was expected due to the way the HFR5-D writes flight data across the six chips. The stream of flight data is striped progressively across each chip. In this case, at 512 data words per second, each chip write consisted of about 1.3 seconds of data. Therefore, the final reconstructed data stream missing chip U2 should appear as chunks of about 6.5 seconds of valid data followed by a 1.3 second gap of missing data from chip U2.

由于缺少芯片映像，在整个数据文件持续时间内出现了规律性的数据丢失。鉴于 HFR5-D 将飞行数据写入六个芯片的方式，这是预期情况。飞行数据流以条带方式逐步写入各个芯片。在本例中，在每秒 512 个数据字的情况下，每次芯片写入包含约 1.3 秒的数据。因此，最终重建且缺少 U2 芯片的数据流应表现为约 6.5 秒有效数据块，随后是来自 U2 芯片的 1.3 秒缺失数据间隙。

When initially reconstructed in the analysis software, the data exhibited much larger gaps than the expected 1.3 seconds, on the order of about 4 seconds. This faulty timing invalidated the time history of the data and made it impossible to align it with CVR events.

在分析软件中初始重建时，数据呈现出的间隙远大于预期的 1.3 秒，约为 4 秒量级。这种错误的定时使数据的时间历程无效，并使其无法与 CVR 事件对齐。

It was found that because the missing data broke across subframe boundaries, the analysis software, expecting to see consecutive subframe sync word markers, padded the data with erroneous extra subframes. These disruptions were caused by the loss of sync whenever the dummy data that replaced chip U2 was processed. The sync loss caused the analysis software to pad 4 full subframes about every 8 subframes (seconds).

经发现，由于缺失数据跨越了子帧边界，分析软件预期看到连续的子帧同步字标记，因此用错误的额外子帧填补了数据。每当处理替代 U2 芯片的虚拟数据时发生同步丢失，从而造成这些中断。同步丢失导致分析软件大约每 8 个子帧（秒）填补 4 个完整子帧。

Correcting the condition of the superfluous subframes was a highly labor-intensive process. To accurately align the data required manually inspecting every gap and subframe boundary in the data stream and hand marking each time the chip U2 expected dropout began and ended and adding in the missing subframe sync word boundary. This allowed the replay software to keep the data stream in sync and display the correct timing for the data.

纠正多余子帧的状况是一个劳动强度极高的过程。要准确对齐数据，需要手动检查数据流中的每一个间隙和子帧边界，并手工标记每次预期的 U2 芯片数据丢失开始和结束的位置，同时加入缺失的子帧同步字边界。这样可使回放软件保持数据流同步，并显示数据的正确定时。

Due to the amount of time it took to correct the timing of the data stream, only the final 12 minutes of the recorded data from the accident flight were corrected to be time accurate. Additionally portions of the previous landing and taxi were corrected to aid in the validation of the individual parameters.

由于纠正数据流定时所需时间较长，仅对事故飞行所记录数据的最后 12 分钟进行了时间准确性校正。此外，还校正了前一次着陆和滑行的部分数据，以帮助验证各个参数。

Parameter validation is the process of evaluating individual parameters to determine whether the parameter was being recorded correctly and accurately by the data system, and check that the scaling and offset of the engineering units are correct. Of the more than 1,000 parameters recorded in this Boeing dataframe, approximately 150 were positively validated. Validation focused on parameters that captured the aircraft's accelerations, position, altitude, attitude, and speed (basic parameters), parameters related to flight controls, surfaces, and flight control forces, and parameters related to the operation of the aircraft's engines. A table of the validated parameters can be found in Appendix B.

参数验证是评估各个参数的过程，用以确定该参数是否由数据系统正确、准确地记录，并检查工程单位的比例系数和偏置是否正确。在该 Boeing dataframe（Boeing 数据帧格式）中记录的 1,000 多个参数里，约 150 个得到了肯定验证。验证重点关注捕获飞机加速度、位置、高度、姿态和速度（基本参数）的参数，与飞行控制、舵面和飞行控制力相关的参数，以及与飞机发动机运行相关的参数。经验证参数表见 Appendix B。

COCKPIT VOICE AND FLIGHT DATA RECORDER COMBINED DOWNLOAD REPORT

驾驶舱语音和飞行数据记录器合并下载报告

DCA22WA102 PG 22 OF 33

DCA22WA102 第 22 页，共 33 页

The FDR data ended with the aircraft still in flight. The data stopped with the aircraft in a descent at approximately 26,000 ft. It did not capture the remainder of the descent and final accident sequence. Investigating the reason for the premature end of the flight data it was found that while at cruise at 29,000 ft, both engine N2 values decreased rapidly below the point at which the generators drop offline. The FDR does not have a battery backup, so without power from the aircraft generators it will power down. This is different from the CVR, which does have a battery backup and can continue recording for at least 10 minutes after the loss of the aircraft generators. Looking for the reason that the engine N2 dropped below the generator cutoff speed, it was found that while cruising at 29,000 ft, the fuel switches on both engines moved from the run position to the cutoff position. Engine speeds decreased after the fuel switch movement.

FDR 数据结束时飞机仍在飞行中。数据在飞机约 26,000 ft 高度下降时停止。它未捕获下降的其余部分和最终事故序列。在调查飞行数据提前结束的原因时发现，当飞机在 29,000 ft 巡航时，两台发动机的 N2 值均迅速下降至低于发电机脱离在线的点。FDR 没有备用电池，因此在没有来自飞机发电机的电力时会关机。这与 CVR 不同，CVR 有备用电池，并且可在飞机发电机失效后继续记录至少 10 分钟。为查找发动机 N2 降至低于发电机切断转速的原因，发现当飞机在 29,000 ft 巡航时，两台发动机的燃油开关均从运行位置移至切断位置。燃油开关移动后，发动机转速下降。

## 2.3 FDR Data Provided to CAAC

## 2.3 提供给 CAAC 的 FDR 数据

Data provided to the CAAC delegation included the following:

提供给 CAAC delegation（CAAC 代表团）的数据包括以下内容：

- All raw chip download image files, including the dummy U2 file used to generate the FDR .dlu files. - All FDR .dlu files generated by the Honeywell CHIPS utility, both directing the utility to skip the erased blocks found in memory and maintain the erased blocks found in memory. - Both methods generated functionally identical files. The skip erased blocks files were used in the NTSB's data analysis tools, as this method most closely models the behavior of the Playback32 utility when creating a .dlu file. - All FDR unpacked binary files generated by the NTSB's CIDER FDR analysis software. - Manually corrected unpacked binary files that correctly time-aligned the final 12 minutes of recorded data. - Plots of validated data, including plots of aircraft basic parameters, flight controls, control forces, and engine parameters. - Plots were generated for time spans of the final 10 minutes of recorded data and the final 100 seconds of recorded data. - Tabular data of all validated parameters in both exact sample timing and timed to the closest 1/16th of a second.

- 所有原始芯片下载映像文件，包括用于生成 FDR .dlu 文件的虚拟 U2 文件。 - 由 Honeywell CHIPS 工具生成的所有 FDR .dlu 文件，包括指示该工具跳过内存中发现的已擦除块以及保留内存中发现的已擦除块两种方式生成的文件。 - 两种方法生成了功能上相同的文件。跳过已擦除块的文件用于 NTSB 的数据分析工具，因为该方法最接近 Playback32 utility（Playback32 工具）在创建 .dlu 文件时的行为模型。 - 由 NTSB 的 CIDER FDR 分析软件生成的所有 FDR 解包二进制文件。 - 经手动校正、可正确时间对齐所记录数据最后 12 分钟的解包二进制文件。 - 经验证数据的图，包括飞机基本参数、飞行控制、控制力和发动机参数图。 - 针对所记录数据最后 10 分钟和所记录数据最后 100 秒的时间跨度生成了图。 - 所有经验证参数的表格数据，包括精确采样定时以及按最接近 1/16th 秒计时的数据。

COCKPIT VOICE AND FLIGHT DATA RECORDER COMBINED DOWNLOAD REPORT DCA22WA102 PG 23 OF 33

驾驶舱语音和飞行数据记录器 合并下载报告 DCA22WA102 第 23 页，共 33 页

- All files necessary for generation of the dataframe in the CAAC's FDR software analysis tools. - All images, scans, and microscopic photographs taken of the FDR board during the data recovery process.

- 在 CAAC 的 FDR 软件分析工具中生成 dataframe 所需的所有文件。 - 数据恢复过程中拍摄的 FDR 板的所有图像、扫描件和显微照片。

The NTSB maintained copies of these FDR files to maintain the ability to assist the CAAC in the development of the flight history and sequence of events for the final report.

NTSB 保留了这些 FDR 文件的副本，以保持协助 CAAC 为最终报告编制飞行历史和事件序列的能力。

Submitted by:

提交人：

Charles Cates Mechanical Engineer/Recorder Specialist

Charles Cates Mechanical Engineer/Recorder Specialist（机械工程师/记录器专家）

COCKPIT VOICE AND FLIGHT DATA RECORDER COMBINED DOWNLOAD REPORT DCA22WA102 PG 24 OF 33

驾驶舱语音和飞行数据记录器 合并下载报告 DCA22WA102 第 24 页，共 33 页

# 2.3.1 FDR Plots

# 2.3.1 FDR 图

China Eastern, Boeing 737-800, MU-5735, B-1791

China Eastern, Boeing 737-800, MU-5735, B-1791

Figure 11. Basic aircraft parameters from final 90 seconds of recording.  Revised: 14 April 2022 National Transportation Safety Board

图 11. 记录最后 90 秒的飞机基本参数。  修订：14 April 2022 National Transportation Safety Board（国家运输安全委员会）

COCKPIT VOICE AND FLIGHT DATA RECORDER

驾驶舱语音和飞行数据记录器

COMBINED DOWNLOAD REPORT

合并下载报告

DCA22WA102

DCA22WA102

PG 25 OF 33

第 25 页，共 33 页

China Eastern, Boeing 737-800, MU-5735, B-1791 Figure 12. Flight control and input positions for final 90 seconds of recording.  Revised: 14 April 2022 National Transportation Safety Board

China Eastern, Boeing 737-800, MU-5735, B-1791 图 12. 记录最后 90 秒的飞行控制和输入位置。  修订：14 April 2022 National Transportation Safety Board

COCKPIT VOICE AND FLIGHT DATA RECORDER

驾驶舱语音和飞行数据记录器

COMBINED DOWNLOAD REPORT

合并下载报告

DCA22WA102

DCA22WA102

PG 26 OF 33

第 26 页，共 33 页

China Eastern, Boeing 737-800, MU-5735, B-1791 Figure 13. Control forces recorded during upset event through end of recording.  Revised: 14 April 2022 National Transportation Safety Board

China Eastern, Boeing 737-800, MU-5735, B-1791 图 13. 在失控事件期间至记录结束时记录的控制力。  修订：14 April 2022 National Transportation Safety Board

COCKPIT VOICE AND FLIGHT DATA RECORDER

驾驶舱语音和飞行数据记录器

COMBINED DOWNLOAD REPORT

合并下载报告

DCA22WA102

DCA22WA102

PG 27 OF 33

第 27 页，共 33 页

# APPENDIX A. CVR QUALITY RATING SCALE

# APPENDIX A. CVR 质量评级尺度

The levels of recording quality are characterized by the following traits of the cockpit voice recorder information:

记录质量等级由以下 cockpit voice recorder（驾驶舱语音记录器）信息特征来表征：

## Excellent Quality

## 优秀质量

Virtually all of the crew conversations could be accurately and easily understood. The transcript that was developed may indicate only one or two words that were not intelligible. Any loss in the transcript is usually attributed to simultaneous cockpit/radio transmissions that obscure each other.

几乎所有机组对话都能够被准确且容易地理解。形成的记录稿可能仅标明一两个无法听清的词。记录稿中的任何缺失通常归因于同时发生的驾驶舱/无线电通信相互遮蔽。

## Good Quality

## 良好质量

Most of the crew conversations could be accurately and easily understood. The transcript that was developed may indicate several words or phrases that were not intelligible. Any loss in the transcript can be attributed to minor technical deficiencies or momentary dropouts in the recording system or to a large number of simultaneous cockpit/radio transmissions that obscure each other.

大多数机组对话都能够被准确且容易地理解。形成的记录稿可能标明若干无法听清的词或短语。记录稿中的任何缺失可归因于轻微技术缺陷、记录系统中的瞬时丢失，或大量同时发生的驾驶舱/无线电通信相互遮蔽。

## Fair Quality

## 一般质量

The majority of the crew conversations were intelligible. The transcript that was developed may indicate passages where conversations were unintelligible or fragmented. This type of recording is usually caused by cockpit noise that obscures portions of the voice signals or by a minor electrical or mechanical failure of the CVR system that distorts or obscures the audio information.

大部分机组对话可以听懂。形成的记录稿可能标明对话无法听清或支离破碎的段落。此类记录通常由遮蔽部分语音信号的驾驶舱噪声造成，或由 CVR 系统轻微的电气或机械故障造成，该故障使音频信息失真或被遮蔽。

## Poor Quality

## 较差质量

Extraordinary means had to be used to make some of the crew conversations intelligible. The transcript that was developed may indicate fragmented phrases and conversations and may indicate extensive passages where conversations were missing or unintelligible. This type of recording is usually caused by a combination of a high cockpit noise level with a low voice signal (poor signal-to-noise ratio) or by a mechanical or electrical failure of the CVR system that severely distorts or obscures the audio information.

必须使用非常规手段才能使部分机组对话可听懂。形成的记录稿可能标明支离破碎的短语和对话，并可能标明对话缺失或无法听清的大段内容。此类记录通常由高驾驶舱噪声水平与低语音信号（差的信噪比）相结合造成，或由 CVR 系统的机械或电气故障造成，该故障严重使音频信息失真或被遮蔽。

## Unusable

## 不可用

Crew conversations may be discerned, but neither ordinary nor extraordinary means made it possible to develop a meaningful transcript of the conversations. This type of recording is usually caused by an almost total mechanical or electrical failure of the CVR system.

可能可分辨出机组对话，但无论使用常规还是非常规手段，都无法形成有意义的对话记录稿。此类记录通常由 CVR 系统几乎完全的机械或电气故障造成。

COCKPIT VOICE AND FLIGHT DATA RECORDER COMBINED DOWNLOAD REPORT DCA22WA102 PG 28 OF 33

驾驶舱语音和飞行数据记录器 合并下载报告 DCA22WA102 第 28 页，共 33 页

# APPENDIX B. VALIDATED PARAMETERS

# 附录 B. 经验证的参数

This appendix describes the parameters validated and provided to the CAAC. Table B-1 lists the parameter names, units, and a description of each parameter. Additionally, Table B-2 describes the units and abbreviations used in this report.

本附录说明经验证并提供给 CAAC（中国民用航空局）的参数。表 B-1 列出了参数名称、单位以及每个参数的说明。此外，表 B-2 说明了本报告中使用的单位和缩写。

Table B-1. Validated and provided FDR parameters

表 B-1. 经验证并提供的 FDR 参数

| No. | Validated Parameter Name | Units | Description | | --- | --- | --- | --- | | 1. | Absolute Roll Rate | deg/s | Aircraft Roll Rate | | 2. | Accel Lat | g | Lateral Acceleration | | 3. | Accel Long | g | Longitudinal Acceleration | | 4. | Accel Vert | g | Vertical Acceleration | | 5. | Active Altitude Ref - FCC | | FCC Altitude Reference | | 6. | Aileron Actuator Pos-L | deg | Left Aileron Actuator Position | | 7. | Aileron Quadrant Pos | deg | Aileron Control Quadrant Position | | 8. | Aileron Roll Cmd-L | deg | Left FCC Aileron Roll Command | | 9. | Aileron-L | deg | Left Aileron Position | | 10. | Aileron-R | deg | Right Aileron Position | | 11. | Air Gnd On Gnd | | Air Ground Sensor on Ground | | 12. | AIR GROUND - SMYDC-1 | | SMYDC 1 On Ground | | 13. | AIR GROUND - SMYDC-2 | | SMYDC 2 On Ground | | 14. | Air-Ground | | Air/Ground | | 15. | Airspeed Comp | kts | Computed Airspeed | | 16. | Airspeed Max Allowable | kts | FCC Max Allowable Airspeed | | 17. | Airspeed Target FCC | kts | FCC Computed Airspeed Target | | 18. | Alt 1 Baro Corr | ft | Barometric Corrected Altitude 1 | | 19. | Alt 2 Baro Corr | ft | Barometric Corrected Altitude 2 | | 20. | Alt 3 Baro Corr | ft | Barometric Corrected Altitude 3 | | 21. | Alt 4 Baro Corr | ft | Barometric Corrected Altitude 4 | | 22. | ALT ACQ Engaged - FCC | | Altitude Acquire Autopilot Mode Engaged | | 23. | Alt Baro Corr Combine | ft | Combined Barometric Corrected Altitude | | 24. | ALT HOLD Engaged - FCC | | Altitude Hold Autopilot Mode Engaged | | 25. | Altitude Press | ft | Pressure Altitude | | 26. | Altitude Radio DEU | ft | Displayed Radio Altitude | | 27. | Altitude Radio-1 | ft | Radio Altimeter 1 | | 28. | Altitude Radio-2 | ft | Radio Altimeter 2 | | 29. | AP Off - FCC | | Autopilot Off | | 30. | AP-1 Warn | | Autopilot Warning 1 | | 31. | AP-2 Warn | | Autopilot Warning 2 | | 32. | APU N1 | %RPM | APU Shaft Speed | | 33. | APU On | | APU Running | | 34. | APU Ready To Load | | APU Ready for Load | | 35. | AT FMC SPD Engaged | | Autothrottle Speed Mode Engaged | | 36. | CMD A - FCC | | FCC A In Command | | 37. | CMD A Light - FCC | | FCC A Light Active | | 38. | CMD B Light - FCC | | FCC B Light Active | | 39. | Ctrl Col Force Pitch CWS | lb | Combined Control Column Force |

| 序号 | 经验证的参数名称 | 单位 | 说明 | | --- | --- | --- | --- | | 1. | Absolute Roll Rate | deg/s | 飞机滚转率 | | 2. | Accel Lat | g | 横向加速度 | | 3. | Accel Long | g | 纵向加速度 | | 4. | Accel Vert | g | 垂直加速度 | | 5. | Active Altitude Ref - FCC | | FCC 高度基准 | | 6. | Aileron Actuator Pos-L | deg | 左副翼作动器位置 | | 7. | Aileron Quadrant Pos | deg | 副翼控制扇形轮位置 | | 8. | Aileron Roll Cmd-L | deg | 左 FCC 副翼滚转指令 | | 9. | Aileron-L | deg | 左副翼位置 | | 10. | Aileron-R | deg | 右副翼位置 | | 11. | Air Gnd On Gnd | | 空/地传感器在地面 | | 12. | AIR GROUND - SMYDC-1 | | SMYDC 1 在地面 | | 13. | AIR GROUND - SMYDC-2 | | SMYDC 2 在地面 | | 14. | Air-Ground | | 空/地 | | 15. | Airspeed Comp | kts | 计算空速 | | 16. | Airspeed Max Allowable | kts | FCC 最大允许空速 | | 17. | Airspeed Target FCC | kts | FCC 计算空速目标 | | 18. | Alt 1 Baro Corr | ft | 气压修正高度 1 | | 19. | Alt 2 Baro Corr | ft | 气压修正高度 2 | | 20. | Alt 3 Baro Corr | ft | 气压修正高度 3 | | 21. | Alt 4 Baro Corr | ft | 气压修正高度 4 | | 22. | ALT ACQ Engaged - FCC | | 高度截获自动驾驶模式接通 | | 23. | Alt Baro Corr Combine | ft | 组合气压修正高度 | | 24. | ALT HOLD Engaged - FCC | | 高度保持自动驾驶模式接通 | | 25. | Altitude Press | ft | 气压高度 | | 26. | Altitude Radio DEU | ft | 显示的无线电高度 | | 27. | Altitude Radio-1 | ft | 无线电高度表 1 | | 28. | Altitude Radio-2 | ft | 无线电高度表 2 | | 29. | AP Off - FCC | | 自动驾驶断开 | | 30. | AP-1 Warn | | 自动驾驶警告 1 | | 31. | AP-2 Warn | | 自动驾驶警告 2 | | 32. | APU N1 | %RPM | APU 轴转速 | | 33. | APU On | | APU 运行中 | | 34. | APU Ready To Load | | APU 准备承载 | | 35. | AT FMC SPD Engaged | | 自动油门速度模式接通 | | 36. | CMD A - FCC | | FCC A 处于指令状态 | | 37. | CMD A Light - FCC | | FCC A 灯点亮 | | 38. | CMD B Light - FCC | | FCC B 灯点亮 | | 39. | Ctrl Col Force Pitch CWS | lb | 组合驾驶杆力 |

COCKPIT VOICE AND FLIGHT DATA RECORDER COMBINED DOWNLOAD REPORT DCA22WA102 PG 29 OF 33

座舱语音和飞行数据记录器 合并下载报告 DCA22WA102 第 29 页，共 33 页

| No. | Validated Parameter Name | Units | Description | | --- | --- | --- | --- | | 40. | Ctrl Col Force Pitch CWS Foreign | lb | Control Column Force Opposite FCC | | 41. | Ctrl Col Force Pitch CWS Local | lb | Control Column Force Commanding FCC | | 42. | Ctrl Col Pos-L | deg | Left Control Column Position | | 43. | Ctrl Col Pos-R | deg | Right Control Column Position | | 44. | Ctrl Whl Force Roll CWS | lb | Control Wheel Force | | 45. | Ctrl Whl Pos-L | deg | Left Control Wheel Position | | 46. | Ctrl Whl Pos-R | deg | Right Control Wheel Position | | 47. | Drift Angle -FMC | deg | Computed Drift Angle | | 48. | Elevator Actuator Pos-L | deg | Left Elevator Actuator Position | | 49. | Elevator Pitch Cmd-L | deg | Left FCC Elevator Pitch Command | | 50. | Elevator-L | deg | Left Elevator Position | | 51. | Elevator-R | deg | Right Elevator Position | | 52. | Eng1 Cutoff SW | | Left Engine Cutoff Switch | | 53. | Eng1 EGT | degC | Left Engine Exhaust Gas Temperature | | 54. | Eng1 Fire | | Left Engine Fire Detected | | 55. | Eng1 FMC N1 Bug Drive | %RPM | Left Engine N1 Bug Drive | | 56. | Eng1 FMC N1 Target | %RPM | Left Engine N1 Target | | 57. | Eng1 FMV Pos | % | Left Engine Fuel Metering Valve Position | | 58. | Eng1 Fuel Flow | pph | Left Engine Fuel Flow | | 59. | Eng1 N1 | %RPM | Left Engine Fan Speed | | 60. | Eng1 N1 Cmd | %RPM | Left Engine Fan Speed Command | | 61. | Eng1 N1 Ref | %RPM | Left Engine Reference Fan Speed | | 62. | Eng1 N1 Tach | %RPM | Left Engine Fan Speed Tach | | 63. | Eng1 N2 Actual | %RPM | Left Engine Core Speed | | 64. | Eng1 N2 Tach | %RPM | Left Engine Core Speed Tach | | 65. | Eng1 Oil Press | psi | Left Engine Oil Pressure | | 66. | Eng1 Oil Qty | qt | Left Engine Oil Quantity | | 67. | Eng1 Oil Temp | degC | Left Engine Oil Temperature | | 68. | Eng1 TRA | deg | Left Engine Throttle Resolver Angle | | 69. | Eng2 Cutoff SW | | Right Engine Cutoff Switch | | 70. | Eng2 EGT | degC | Right Engine Exhaust Gas Temperature | | 71. | Eng2 Fire | | Right Engine Fire Detected | | 72. | Eng2 FMC N1 Bug Drive | %RPM | Right Engine N1 Bug Drive | | 73. | Eng2 FMC N1 Target | %RPM | Right Engine N1 Target | | 74. | Eng2 FMV Pos | % | Right Engine Fuel Metering Valve Position | | 75. | Eng2 Fuel Flow | pph | Right Engine Fuel Flow | | 76. | Eng2 N1 | %RPM | Right Engine Fan Speed | | 77. | Eng2 N1 Cmd | %RPM | Right Engine Fan Speed Command | | 78. | Eng2 N1 Ref | %RPM | Right Engine Reference Fan Speed | | 79. | Eng2 N1 Tach | %RPM | Right Engine Fan Speed Tach | | 80. | Eng2 N2 Actual | %RPM | Right Engine Core Speed | | 81. | Eng2 N2 Tach | %RPM | Right Engine Core Speed Tach | | 82. | Eng2 Oil Press | psi | Right Engine Oil Pressure | | 83. | Eng2 Oil Qty | qt | Right Engine Oil Quantity | | 84. | Eng2 Oil Temp | degC | Right Engine Oil Temperature | | 85. | Eng2 TRA | deg | Right Engine Throttle Resolver Angle | | 86. | FAC Engage - FCC | | Final Approach Course AP Mode Engaged |

| 序号 | 经验证的参数名称 | 单位 | 说明 | | --- | --- | --- | --- | | 40. | Ctrl Col Force Pitch CWS Foreign | lb | 与 FCC 相对的驾驶杆力 | | 41. | Ctrl Col Force Pitch CWS Local | lb | 指令 FCC 的驾驶杆力 | | 42. | Ctrl Col Pos-L | deg | 左驾驶杆位置 | | 43. | Ctrl Col Pos-R | deg | 右驾驶杆位置 | | 44. | Ctrl Whl Force Roll CWS | lb | 驾驶盘力 | | 45. | Ctrl Whl Pos-L | deg | 左驾驶盘位置 | | 46. | Ctrl Whl Pos-R | deg | 右驾驶盘位置 | | 47. | Drift Angle -FMC | deg | 计算偏流角 | | 48. | Elevator Actuator Pos-L | deg | 左升降舵作动器位置 | | 49. | Elevator Pitch Cmd-L | deg | 左 FCC 升降舵俯仰指令 | | 50. | Elevator-L | deg | 左升降舵位置 | | 51. | Elevator-R | deg | 右升降舵位置 | | 52. | Eng1 Cutoff SW | | 左发动机切断电门 | | 53. | Eng1 EGT | degC | 左发动机排气温度 | | 54. | Eng1 Fire | | 探测到左发动机火警 | | 55. | Eng1 FMC N1 Bug Drive | %RPM | 左发动机 N1 游标驱动 | | 56. | Eng1 FMC N1 Target | %RPM | 左发动机 N1 目标 | | 57. | Eng1 FMV Pos | % | 左发动机燃油计量活门位置 | | 58. | Eng1 Fuel Flow | pph | 左发动机燃油流量 | | 59. | Eng1 N1 | %RPM | 左发动机风扇转速 | | 60. | Eng1 N1 Cmd | %RPM | 左发动机风扇转速指令 | | 61. | Eng1 N1 Ref | %RPM | 左发动机基准风扇转速 | | 62. | Eng1 N1 Tach | %RPM | 左发动机风扇转速表读数 | | 63. | Eng1 N2 Actual | %RPM | 左发动机核心转速 | | 64. | Eng1 N2 Tach | %RPM | 左发动机核心转速表读数 | | 65. | Eng1 Oil Press | psi | 左发动机滑油压力 | | 66. | Eng1 Oil Qty | qt | 左发动机滑油量 | | 67. | Eng1 Oil Temp | degC | 左发动机滑油温度 | | 68. | Eng1 TRA | deg | 左发动机油门解算器角度 | | 69. | Eng2 Cutoff SW | | 右发动机切断电门 | | 70. | Eng2 EGT | degC | 右发动机排气温度 | | 71. | Eng2 Fire | | 探测到右发动机火警 | | 72. | Eng2 FMC N1 Bug Drive | %RPM | 右发动机 N1 游标驱动 | | 73. | Eng2 FMC N1 Target | %RPM | 右发动机 N1 目标 | | 74. | Eng2 FMV Pos | % | 右发动机燃油计量活门位置 | | 75. | Eng2 Fuel Flow | pph | 右发动机燃油流量 | | 76. | Eng2 N1 | %RPM | 右发动机风扇转速 | | 77. | Eng2 N1 Cmd | %RPM | 右发动机风扇转速指令 | | 78. | Eng2 N1 Ref | %RPM | 右发动机基准风扇转速 | | 79. | Eng2 N1 Tach | %RPM | 右发动机风扇转速表读数 | | 80. | Eng2 N2 Actual | %RPM | 右发动机核心转速 | | 81. | Eng2 N2 Tach | %RPM | 右发动机核心转速表读数 | | 82. | Eng2 Oil Press | psi | 右发动机滑油压力 | | 83. | Eng2 Oil Qty | qt | 右发动机滑油量 | | 84. | Eng2 Oil Temp | degC | 右发动机滑油温度 | | 85. | Eng2 TRA | deg | 右发动机油门解算器角度 | | 86. | FAC Engage - FCC | | 最后进近航道 AP 模式接通 |

COCKPIT VOICE AND FLIGHT DATA RECORDER COMBINED DOWNLOAD REPORT DCA22WA102 PG 30 OF 33

座舱语音和飞行数据记录器 合并下载报告 DCA22WA102 第 30 页，共 33 页

| No. | Validated Parameter Name | Units | Description | | --- | --- | --- | --- | | 87. | FCC In Command of MACH Trim - R | | Right FCC Commanding Mach Trim | | 88. | FCC-L In Command of MACH Trim | | Left FCC Commanding Mach Trim | | 89. | FD-A Switch - FCC | | Flight Director A Switch | | 90. | FD-B Switch - FCC | | Flight Director B Switch | | 91. | Flap Handle Pos | deg | Flap Handle Position | | 92. | Flap-L | deg | Left Flap Position | | 93. | Flap-R | deg | Right Flap Position | | 94. | FMC Selected Airspeed | kts | Selected Airspeed - FMC | | 95. | FMC Selected Altitude | ft | Selected Altitude - FMC | | 96. | FMC Selected Mach | mach | Selected Mach Number - FMC | | 97. | FMC Valid | | FMC Valid | | 98. | G/S Dev Warn - FCC | | Glideslope Deviation Warning | | 99. | GP Engage - FCC | | Glidepath AP Mode Engaged | | 100. | Ground Spd | kts | Ground Speed | | 101. | Groundspeed FMC | kts | Ground Speed Calculated by FMC | | 102. | Groundspeed Disp -L | kts | Ground Speed Displayed on Left PFD | | 103. | GS Engaged - FCC | | Glideslope AP Mode Engaged | | 104. | HDG SEL Light - FCC | | Heading Select Light Active | | 105. | HDG SELECT - FCC | | Heading Select AP Mode Engaged | | 106. | Heading | deg | Magnetic Heading | | 107. | Heading Selected FCC | deg | FCC Heading Selected | | 108. | High Speed Buffet Speed | kts | High Speed Buffet Speed | | 109. | Hyd Oil Press - A | psi | Hydraulic Pressure System A | | 110. | Hyd Oil Press - B | psi | Hydraulic Pressure System B | | 111. | Hyd Oil Qty - A | % | Hydraulic Oil Quantity System A | | 112. | Hyd Oil Qty - B | % | Hydraulic Oil Quantity System B | | 113. | Hydraulic Oil Pressure Standby | psi | Standby Hydraulic Pressure | | 114. | Hydraulic System A ELEC | | Hydraulic A Electric Pump | | 115. | Hydraulic System A Eng 1 | | Hydraulic A Engine Pump | | 116. | Hydraulic System B ELEC | | Hydraulic B Electric Pump | | 117. | Hydraulic System B Eng 2 | | Hydraulic B Engine Pump | | 118. | Hydraulic System Standby | | Hydraulic Standby System Engaged | | 119. | IAS Display - FCC | | Indicated Airspeed Displayed | | 120. | LNAV Engaged - FCC | | LNAV AP Mode Engaged | | 121. | LNAV Light - FCC | | LNAV Light Active | | 122. | LOC Engaged - FCC | | Localizer AP Mode Engaged | | 123. | LOCAL LIMITED MASTER FCC-L | | FCC Left Master | | 124. | LOCAL LIMITED MASTER FCC-R | | FCC Right Master | | 125. | LVL Change Light - FCC | | Flight Level Change AP Mode Active | | 126. | MACH Trim Servo Brake Status - FCC-L | | Mach Trim Servo Status | | 127. | N1 Light - FCC | | N1 Light Active | | 128. | N1 Limit Mode Cmd - FCC | | N1 Limit AT Mode Engaged | | 129. | Pitch Angle | deg | Aircraft Pitch Angle | | 130. | Roll Angle | deg | Aircraft Roll Angle | | 131. | Roll Rate | deg/s | Aircraft Roll Rate | | 132. | Rudder | deg | Rudder Position | | 133. | Rudder Ped Pos | deg | Rudder Pedal Position |

| 序号 | 经验证的参数名称 | 单位 | 说明 | | --- | --- | --- | --- | | 87. | FCC In Command of MACH Trim - R | | 右 FCC 指令 Mach 配平 | | 88. | FCC-L In Command of MACH Trim | | 左 FCC 指令 Mach 配平 | | 89. | FD-A Switch - FCC | | 飞行指引仪 A 电门 | | 90. | FD-B Switch - FCC | | 飞行指引仪 B 电门 | | 91. | Flap Handle Pos | deg | 襟翼手柄位置 | | 92. | Flap-L | deg | 左襟翼位置 | | 93. | Flap-R | deg | 右襟翼位置 | | 94. | FMC Selected Airspeed | kts | 选定空速 - FMC | | 95. | FMC Selected Altitude | ft | 选定高度 - FMC | | 96. | FMC Selected Mach | mach | 选定 Mach 数 - FMC | | 97. | FMC Valid | | FMC 有效 | | 98. | G/S Dev Warn - FCC | | 下滑道偏差警告 | | 99. | GP Engage - FCC | | 下滑路径 AP 模式接通 | | 100. | Ground Spd | kts | 地速 | | 101. | Groundspeed FMC | kts | FMC 计算的地速 | | 102. | Groundspeed Disp -L | kts | 左 PFD 显示的地速 | | 103. | GS Engaged - FCC | | 下滑道 AP 模式接通 | | 104. | HDG SEL Light - FCC | | 航向选择灯点亮 | | 105. | HDG SELECT - FCC | | 航向选择 AP 模式接通 | | 106. | Heading | deg | 磁航向 | | 107. | Heading Selected FCC | deg | FCC 选定航向 | | 108. | High Speed Buffet Speed | kts | 高速抖振速度 | | 109. | Hyd Oil Press - A | psi | A 系统液压压力 | | 110. | Hyd Oil Press - B | psi | B 系统液压压力 | | 111. | Hyd Oil Qty - A | % | A 系统液压油量 | | 112. | Hyd Oil Qty - B | % | B 系统液压油量 | | 113. | Hydraulic Oil Pressure Standby | psi | 备用液压压力 | | 114. | Hydraulic System A ELEC | | 液压 A 电动泵 | | 115. | Hydraulic System A Eng 1 | | 液压 A 发动机泵 | | 116. | Hydraulic System B ELEC | | 液压 B 电动泵 | | 117. | Hydraulic System B Eng 2 | | 液压 B 发动机泵 | | 118. | Hydraulic System Standby | | 备用液压系统接通 | | 119. | IAS Display - FCC | | 显示的指示空速 | | 120. | LNAV Engaged - FCC | | LNAV AP 模式接通 | | 121. | LNAV Light - FCC | | LNAV 灯点亮 | | 122. | LOC Engaged - FCC | | 航向道 AP 模式接通 | | 123. | LOCAL LIMITED MASTER FCC-L | | FCC 左主控 | | 124. | LOCAL LIMITED MASTER FCC-R | | FCC 右主控 | | 125. | LVL Change Light - FCC | | 飞行高度层改变 AP 模式启用 | | 126. | MACH Trim Servo Brake Status - FCC-L | | Mach 配平伺服状态 | | 127. | N1 Light - FCC | | N1 灯点亮 | | 128. | N1 Limit Mode Cmd - FCC | | N1 限制 AT 模式接通 | | 129. | Pitch Angle | deg | 飞机俯仰角 | | 130. | Roll Angle | deg | 飞机滚转角 | | 131. | Roll Rate | deg/s | 飞机滚转率 | | 132. | Rudder | deg | 方向舵位置 | | 133. | Rudder Ped Pos | deg | 方向舵脚蹬位置 |

COCKPIT VOICE AND FLIGHT DATA RECORDER COMBINED DOWNLOAD REPORT DCA22WA102 PG 31 OF 33

座舱语音和飞行数据记录器 合并下载报告 DCA22WA102 第 31 页，共 33 页

| No. | Validated Parameter Name | Units | Description | | --- | --- | --- | --- | | 134. | Rudder Pos- LVDT DEMOD-STBY PCU | deg | Rudder Standby PCU LVDT Position | | 135. | Rudder Servo Cmd-STBY PCU | deg | Rudder Standby PCU Servo Command | | 136. | Selected Airspeed FCC | kts | Selected Airspeed - FCC | | 137. | Selected Altitude FCC | ft | Selected Altitude - FCC | | 138. | Selected Course Foreign FCC | deg | Selected Course Foreign FCC | | 139. | Selected Course Local FCC | deg | Selected Course Local FCC | | 140. | Selected Mach FCC | mach | Selected Mach Number - FCC | | 141. | Selected Vertical Speed FCC | fpm | Selected Vertical Speed - FCC | | 142. | Single Channel - FCC | | AP on Single Channel | | 143. | SPD Light On - FCC | | SPD Light Active | | 144. | SPEED INTERVENTION ACTIVE - FCC | | Speed Intervention Active | | 145. | TOGA Engaged - FCC | | AT TOGA Mode Engaged | | 146. | Track Angle True FMC | deg | True Track Angle | | 147. | VISUAL ALTITUDE ALERT - FCC | | Visual Altitude Alert Active | | 148. | VNAV Light On - FCC | | VNAV Light Active | | 149. | VNAV PATH Engaged - FCC | | VNAV Path AP Mode Engaged | | 150. | VNAV SPD Engaged - FCC | | VNAV Spd AP Mode Engaged | | 151. | Wind Direction True -FMC | deg | Wind Direction | | 152. | Wind Speed -FMC | kts | Wind Speed | | 153. | Yaw Rate | deg/s | Aircraft Yaw Rate |

| 序号 | 经验证的参数名称 | 单位 | 说明 | | --- | --- | --- | --- | | 134. | Rudder Pos- LVDT DEMOD-STBY PCU | deg | 方向舵备用 PCU LVDT 位置 | | 135. | Rudder Servo Cmd-STBY PCU | deg | 方向舵备用 PCU 伺服指令 | | 136. | Selected Airspeed FCC | kts | 选定空速 - FCC | | 137. | Selected Altitude FCC | ft | 选定高度 - FCC | | 138. | Selected Course Foreign FCC | deg | 对侧 FCC 选定航道 | | 139. | Selected Course Local FCC | deg | 本侧 FCC 选定航道 | | 140. | Selected Mach FCC | mach | 选定 Mach 数 - FCC | | 141. | Selected Vertical Speed FCC | fpm | 选定垂直速度 - FCC | | 142. | Single Channel - FCC | | AP 单通道工作 | | 143. | SPD Light On - FCC | | SPD 灯点亮 | | 144. | SPEED INTERVENTION ACTIVE - FCC | | 速度干预启用 | | 145. | TOGA Engaged - FCC | | AT TOGA 模式接通 | | 146. | Track Angle True FMC | deg | 真航迹角 | | 147. | VISUAL ALTITUDE ALERT - FCC | | 目视高度警戒启用 | | 148. | VNAV Light On - FCC | | VNAV 灯点亮 | | 149. | VNAV PATH Engaged - FCC | | VNAV 路径 AP 模式接通 | | 150. | VNAV SPD Engaged - FCC | | VNAV 速度 AP 模式接通 | | 151. | Wind Direction True -FMC | deg | 风向 | | 152. | Wind Speed -FMC | kts | 风速 | | 153. | Yaw Rate | deg/s | 飞机偏航率 |

Note: This FDR records pressure altitude, which is based on a standard altimeter setting of 29.92 inches of mercury (in Hg). The pressure altitude information presented in the FDR plots and in the tabular data has not been corrected for the local altimeter setting at the time of the event.

注：该 FDR 记录气压高度，该高度基于 29.92 inches of mercury (in Hg) 的标准高度表设定。FDR 图表和表格数据中呈现的气压高度信息未根据事件发生时的当地高度表设定进行修正。

Note: Parameters with a blank unit description in table B-1 are discretes. A discrete is typically a 1-bit parameter that is either a 0 state or a 1 state where each state is uniquely defined for each parameter.

注：表 B-1 中单位说明为空白的参数为离散量。离散量通常是 1-bit 参数，为 0 状态或 1 状态，且每个状态均针对每个参数作唯一限定。

COCKPIT VOICE AND FLIGHT DATA RECORDER COMBINED DOWNLOAD REPORT DCA22WA102 PG 32 OF 33

座舱语音和飞行数据记录器 合并下载报告 DCA22WA102 第 32 页，共 33 页

Table B-2. Units and abbreviations

表 B-2. 单位和缩写

| Unit/Abbreviation | Description | | --- | --- | | Alt | Altitude | | AP | Autopilot | | APU | Auxiliary Power Unit | | AT | Autothrottle | | Baro | Barometric | | CWS | Control Wheel Steering | | deg | degrees | | degC | degrees Celsius | | FCC | Flight Control Computer | | FMC | Flight Management Computer | | fpm | feet per minute | | ft | feet | | kts | knots | | lb | pounds | | LVDT | Linear Variable Differential Transformer | | MCP | Mode Control Panel | | PCU | Power Control Unit | | PFD | Primary Flight Display | | pph | pounds per hour | | psi | pounds per square inch | | qt | quart | | RPM | revolutions per minute | | s | second | | SMYDC | Stall Management and Yaw Damper Computer |

| 单位/缩写 | 说明 | | --- | --- | | Alt | 高度 | | AP | 自动驾驶 | | APU | 辅助动力装置 | | AT | 自动油门 | | Baro | 气压 | | CWS | 驾驶盘操纵 | | deg | 度 | | degC | 摄氏度 | | FCC | 飞行控制计算机 | | FMC | 飞行管理计算机 | | fpm | 英尺每分钟 | | ft | 英尺 | | kts | 节 | | lb | 磅 | | LVDT | 线性可变差动变压器 | | MCP | 模式控制面板 | | PCU | 动力控制组件 | | PFD | 主飞行显示器 | | pph | 磅每小时 | | psi | 磅每平方英寸 | | qt | 夸脱 | | RPM | 转每分钟 | | s | 秒 | | SMYDC | 失速管理与偏航阻尼计算机 |

COCKPIT VOICE AND FLIGHT DATA RECORDER COMBINED DOWNLOAD REPORT DCA22WA102 PG 33 OF 33

座舱语音和飞行数据记录器 合并下载报告 DCA22WA102 第 33 页，共 33 页