曝光臺 注意防騙 網曝天貓店富美金盛家居專營店坑蒙拐騙欺詐消費者

was manually pulled. The snap sound was identifiable on the ground without engines and
air-conditioning operating. However in the flight tests, the addition of the background
cockpit noise present during normal cruise obscures the sounds associated with the
manual in-flight pulling of the cockpit circuit breaker. No corresponding sound signatures
could be found in the accident recording.
The summaries of the results of the second tests are as follows:
• During an overload and a short circuit, the sound of the circuit breaker popping was
loud enough to be identified on the CVR‘s area microphone channel, both on the
ground and in-flight.
• During an overload and a short circuit, the CVR recorded unique and identifiable
sound signature on one or more of the channels, both on the ground and in-flight.
• During the manual pull test on the ground, the sound of the circuit breaker was loud
enough to be identified on the CVR recording.
• In cruise flight, normal cockpit background noise obscured the manual circuit breaker
pull sounds. There were no unique electronic identifying sound signature recorded on
the CVR.
The third test
The test was conducted in-flight using a B-737 SilkAir sister aircraft in Singapore on 16
October 1998 and supervised by the Indonesian AAIC (now NTSC), a FAA avionics
inspector (representing NTSB) and Singapore MCIT representatives.
In the third test, several scenarios were performed where the CVR circuit breaker in the
cockpit was manually pulled. The manual pulls were categorized as “soft”, “hard” and
“string” pull. The soft pull was by pulling the circuit breaker with minimum noise. The
hard pull was by pulling the circuit breaker normally. The string pull was by pulling on a
string that was attached to the circuit breaker. This was to simulate a short circuit causing
the circuit breaker to pop out.
All the tests were conducted with an identical AlliedSignal SSCVR 2-hours recorder as
installed in the accident aircraft.
All four channels of the CVR recordings of the above three tests were analyzed using the
same NTSB signal processing software that was used to analyze the accident CVR
recording.
Several tests were done to document the sound that were recorded on the CVR during a
soft, hard and string pull of the CVR circuit breaker. The test closely matched the data
obtained from the second test (NTSB in-flight test above).
1.16.2 Captain Seat Belt Buckle Sound Test
To further understand several “metallic snap/clunk” sounds heard during the last few
seconds of the accident recording, several seatbelt unbuckling actions were performed
20
during the in-flight test on 16 October 1998. The test was done with the PIC seatbelt
buckle only as the PIC had indicated his intention to leave the cockpit. The seatbelt
sounds recorded during this test were then compared with the sounds from the accident
recording.
The accident aircraft was equipped with headsets with “hot” boom microphones, in
addition to the cockpit area microphone. In the MI 185 recording, the “metallic snap”
sounds were picked up on the cockpit area microphone (Channel 4) of the CVR. The
sounds were also picked up by the Captain’s boom microphone (Channel 1) and the First
Officer’s boom microphone (Channel 2) of the CVR recording. The observer’s station
(Channel 3) contained no cockpit audio information probably because no microphone was
attached.
The tests showed that the metallic snap sound associated with the Captain’s right seatbelt
buckle was of sufficient intensity to be recorded in all three channels (1, 2 and 4) of the
CVR. The sound associated with the Captain’s left seatbelt buckle was picked up by the
cockpit area microphone (Channel 4) and the First Officer’s boom microphone (Channel
2) but not by the Captain’s boom microphone (Channel 1).
1.16.3 Voice Recognition of ATC Recording using Audio Spectral Analysis
According to FDR and radar data, MI 185 was still flying at FL350 for 6 minutes and 56
seconds after the CVR stopped recording. When the Jakarta ATC recording was
compared to the CVR recording, it was noted that there was one routine radio
transmission from MI 185 after the stoppage of the CVR. This was also confirmed from
the VHF keying data of the FDR recording.
To confirm who made the last MI 185 transmission, i.e. “SilkAir one eight five roger …”
at 09:10:26, a test was performed by comparing words from the last transmission in the
ATC recording with the same words from the CVR recording from an earlier part of the
flight. The First Officer made most of the air to ground radio transmissions. The flight’s
　
中國航空網 m.k6050.com
航空翻譯 www.aviation.cn
本文鏈接地址：NTSC Aircraft Accident Report SILKAIR FLIGHT MI 185 BOEING B(16)