- 30 -
1.14.3.4.3
The tape transport was relatively intact, although the top cover plate of the reels had been detached and the upper reel removed. In addition, a number of the wires to the tape heads had been disconnected. There was evidence of salt water corrosion.
1.14.3.5
1.14.3.5.1
tape transport.
1.14.3.5.2
Inspection of the DFDR tape
The DFDR tape was handed over to ICAO wound on a reel of the type missing from the
On the first 11.8 inches of the Vicalloy tape, the side that would have been adjacent to the hub of the reel showed a white deposit that appeared to be the result of salt water corrosion. The length of tape affected by the deposit was consistent with the length that would have been in contact with the light alloy hub.
1.14.3.5.3 A short distance along the tape there were three holes, each 2.0 inches apart, used to trigger tape reversal. The garnet film technique was used to determine on which side of the tape the recordings had been made. The magnetic patterns were noticeably sharper on the inside face of the tape. This was consistent with the configuration of the tape in the DFDR. The garnet film technique was also used to determine the width and spacing of the four tracks of data recorded on the tape. These were found to agree with the nominal values to within an acceptable degree of tolerance.
1.14.3.5.4 The full length of the tape was subject to examination. Spliced joints were found at approximately 108, 440, 442 and 463 ft from the beginning of the tape. The middle two were spaced at a distance corresponding to the length of the tape between the two reels and the last data was recorded between these two joints. It was not unusual for the tape to break as a result of high speed impacts, near where it left the reels.
1.14.3.6
1.14.3.6.1
DFDR tape playback
To provide maximum capability to validate the information from the last flight, all of the data recorded during the previous twenty-five hours was studied.
1.14.3.6.2 At the time of missile detonation, a brief loss in synchronization lasting 1.4 seconds occurred at the first indication of abnormal measurements. Further losses occurred over a 14 second period about 37 seconds later. For each of these losses, bit dumps of the serial digital data were obtained and edited manually to recover the majority of the lost measurements. These were then incorporated into the engineering unit data files that were used for analysis and-plotting of graphs.
1.14.3.7
Flights recorded on the DFDR
1.14.3.7.1 The oldest recorded data commenced during an easterly flight 6.5 hours prior to a landing at Anchorage on a flight which originated in Seoul as the first leg of flight KE 008 on 30 August 1983. This was followed by the second leg of this flight from Anchorage to JFK Airport in New York on the same date. The next data recorded was for the first leg of flight KE 007 from JFK to Anchorage on 31 August 1983, and the last data was for the flight from Anchorage to the end of recording, approximately 5.5 hours after take-off. The record of these flights constituted approximately 27 hours of recording.