Eastern Air Lines 401

• The flight was uneventful until the approach to MIA. The landing gear handle was placed in the "down" position during the preparation for landing, and the green light, which would have indicated to the flight crew that the nose landing gear was fully extended and locked, failed to illuminate. The captain recycled the landing gear, but the green light still failed to illuminate.
• At 2336:04, the captain instructed the first officer, who was flying the aircraft, to engage the autopilot. The first officer acknowledged the instruction.
• The first officer successfully removed the nose gear light lens assembly, but it jammed when he attempted to replace it.
• At 2337:08, the captain instructed the second officer to enter the forward electronics bay, below the flight deck, to check visually the alignment of the nose gear indices.
• Meanwhile, the flight crew continued their attempts to free the nose gear position light lens from its retainer, without success.
• From 2338:56 until 2341:05, the captain and the first officer discussed the faulty nose gear position light lens assembly and how it might have been reinserted incorrectly.
• At 2340:38, a half-second C-chord, which indicated a deviation of ± 250 feet from the selected altitude, sounded in the cockpit. No crewmember commented on the C-chord. No pitch change to correct for the loss of altitude was recorded.
• Shortly after 2341, the second officer raised his head into the cockpit and stated, "I can't see it, it's pitch dark and I throw the little light, I get, ah, nothing."
• The flight crew and an Eastern Air Lines maintenance specialist who was occupying the forward observer seat then discussed the operation of the nose wheel well light. Afterward, the specialist went into the electronics bay to assist the second officer.
• At 2341:40, MIA approach control asked, "Eastern, ah, four oh one how are things comin' along out there?"
• This query was made a few seconds after the MIA controller noted an altitude reading of 900 feet in the EAL 401 alphanumeric data block on his radar display. The controller testified that he contacted EAL 401 because the flight was nearing the airspace boundary within his jurisdiction. He further stated that he had no doubt at that moment about the safety of the aircraft. Momentary deviations in altitude information on the radar display, he said, are not uncommon; and more than one scan on the display would be required to verify a deviation requiring controller action.

Source: NTSB AAR-73-14, ¶1.1.

• At 2341:44, EAL 401 replied to the controller's query with, "Okay, we'd like to turn around and come, come back in," and at 2341:47, approach control granted the request with "Eastern four oh one turn left heading one eight zero." EAL 401 acknowledged and started the turn.
• At 2342:05, the first officer said, "We did something to the altitude." The captain's reply was, "What?"
• At 2342:07, the first officer asked, "We're still at two thousand, right?" and the captain immediately exclaimed, "Hey, what's happening here?"
• At 2342:10, the first of six radio altimeter warning "beep" sounds began; they ceased immediately before the sound of the initial ground impact.
• At 2342:12, while the aircraft was in a left bank of 28°, it crashed into the Everglades at a point 18.7 statute miles west-northwest of MIA (latitude 25°52' N., longitude 80°36' W.). The aircraft was destroyed by the impact.
• Local weather at the time of the accident was clear, with unrestricted visibility. The accident occurred in darkness, and there was no Moon.

Source: NTSB AAR-73-14, ¶1.1.

• The two autopilot-engage switches and the two flight director system select switches were found in the "off" position. An altitude of 2,000 feet was found selected in the altitude select window.
• The autopilot was engaged at various times during the flight, and was in the control wheel steering (CWS) pitch mode during the last 288 seconds of the flight.
• The autopilot, when engaged in a command mode of operation, will provide total control of the aircraft in accordance with selected heading, pitch, or navigational system inputs. In this mode of operation, the autopilot signals are derived from various computers and sensors in the integrated avionics flight control system.
• When operating in any mode, the selected heading or pitch command function may be disengaged by an overriding 15-pound force applied to the respective, i.e., lateral or pitch, control system through the control wheel. If the force is applied to the pitch control system, only pitch axis control will be effected, reverting to the basic attitude stabilization mode of operation. If the force is applied to the roll control system, the autopilot engage lever will revert to the CWS position.
• The altitude hold mode of operation is unique in that, although it is a command function, it may be engaged when the autopilot is selected to provide either basic CWS or Command operation. When altitude hold is selected, the autopilot provides pitch signals to maintain the altitude existing at the time of engagement. As described, pilot-applied pitch forces on the control wheel will cause disengagement of the altitude hold function, reverting the autopilot pitch channel to attitude stabilization sensitive to control wheel inputs. The autopilot engagement lever will, however, remain in the previously selected position, i. e., either CWS or command. It is possible, therefore, to disengage altitude hold without an accompanying "CMD DISC" warning appearing on the captain or first officer annunciator panels. The normal indications of such an occurrence would be only the extinguishing of the altitude mode select light on the glare shield and the disappearance of the "ALT" annunciation on both annunciator panels.
• The two pitch computers in N310EA were not matched. The pitch override force required to disengage the altitude hold function in computer "A" was 15 pounds, whereas in computer "B" it was 20 pounds. As a result of the mismatch, it would be possible, with the "A" autopilot system engaged, to disengage the "A" AFCS computer, but not the "B" AFCS computer. In this situation, the altitude mode select light would remain on, the "ALT" indication on the captain's annunciator panel would go out, and the same indication on the first officer's annunciator panel would remain on, which would give the first officer the erroneous indication that the autopilot was engaged in the altitude hold mode.
• Because of this mismatch and the system design, a force exerted on the captain's control wheel in excess of 15 pounds, but less than 20 pounds, could result in disengagement of the altitude hold function without the occurrence of a corresponding indication of the first officer's annunciator panel. This would lead to a situation in which the first officer, unaware that altitude hold had been disengaged, would not be alerted to the aircraft altitude deviation. If the autopilot system "B" was engaged, as is believed to have happened, such a situation could not have occurred since a force in excess of 20 pounds would have been required to disengage the altitude hold function and both annunciator panels would have indicated correctly. Therefore, the Board concludes that the mismatched pitch computers in the autoflight system were not a critical factor in this accident.

Source: NTSB AAR-73-14, ¶1

• However, it is significant that recognition of the aforementioned 100-foot loss took 30 seconds after the 0.04 g pitch transient occurred, and after a heading change was requested by approach control. The DFDR readout indicates a 0.9° pitch up maneuver coincident with a change of heading. It was concluded from the DFDR analysis of lateral control system motions that the heading select mode was used for the last 255 seconds of flight to control the aircraft to a heading of 270°. Since selection of the new heading would have required action by the first officer, which included attention to the autopilot control panel, it is reasonable to assume that he should have been aware of the selected heading select functions at this time. It is also reasonable to assume that the autopilot was set up to provide pitch attitude stabilization sensitive to control wheel inputs and heading select, wherein lateral guidance signals were provided to achieve and maintain the 270° heading.
• A series of reductions in power began 16O seconds before impact. The power reductions and slight nose down pitch control movements together were responsible for the unrecognized descent which followed.
• An indication that the throttles were not retarded by a properly operating autothrottle system is the sequence in which the power was reduced. The first power reduction occurred on the Nos. 2 and 3 engines 160 seconds before impact. In the second reduction, the power on the No. 1 engine was matched with the power on the Nos. 2 and 3 engines. Finally, the power on the No. 1 engine was retarded for more than 10 seconds before reduction of power in the two other engines. The throttles were clutched together and driven simultaneously by one servo. If the autothrottle system was "on," only intermittent and random failures in the clutch system would have produced asymmetrical reduction of power similar to that typical of manual throttle movement. Since the autothrottle system of N310EA was found to have been functional, the Board does not believe that this system was involved in the reduction of thrust.
• Regardless of the way in which the status of the autoflight system was indicated to the flight crew, or the manner in which the thrust reduction occurred, the flight instruments (altimeters, vertical speed indicators, airspeed indicators, pitch attitude indicators, and the autopilot vertical speed selector) would have indicated abnormally for a level-flight condition. Together with the altitude-alerting, l/2-second, C-chord signal, the flight instrument indications should have alerted the crew to the undesired descent.

Source: NTSB AAR-73-14, ¶1

• At approximately 2337, some 288 seconds prior to impact, the DFDR readout indicates a vertical acceleration transient of 0. 04 g causing a 200-f.p.m. rate of descent. For a pilot to induce such a transient, he would have to intentionally or inadvertently disengage the altitude hold function. It is conceivable that such a transient could have been produced by an inadvertent action on the part of the pilot which caused a force to be applied to the control column. Such a force would have been sufficient to disengage the altitude hold mode. It was noted that the pitch transient occurred at the same time the captain commented to the second officer to "Get down there and see if the . . . nose wheel's down." If the captain had applied a force to the control wheel while turning to talk to the second officer, the altitude hold function might have been accidentally disengaged.
• Reference to the DFDR shows that power on the No. 3 engine was increased slightly, 1 minute before reduction of power on the Nos. 2 and 3 engines (the initiation of the descent profile). This is a normal manual adjustment typically made by a pilot, and cannot be accomplished by the autothrottle system. Additionally, the speed found set on the autothrottle selector dial was 160 knots, a speed well below that attained or maintained during the last 4 minutes of flight.

Source: NTSB AAR-73-14, ¶2.1

• The three flight crewmembers were preoccupied in an attempt to ascertain the position of the nose landing gear.
• The flight crew did not hear the aural altitude alert which sounded as the aircraft descended through 1,750 feet m.s.l.
• There were several manual thrust reductions during the final descent.
• The flight crew did not monitor the flight instruments, during the final descent until seconds before impact.
• The captain failed to assure that a pilot was monitoring the progress of the aircraft at all times.

Source: NTSB AAR-73-14, ¶3.1

The National Transportation Safety Board determines that the probable cause of this accident was the failure of the flight crew to monitor the flight instruments during the final 4 minutes of flight, and to detect an unexpected descent soon enough to prevent impact with the ground. Preoccupation with a malfunction of, the nose landing gear position indicating system distracted the crew's attention from the instruments and allowed the descent to go unnoticed.

Source: NTSB AAR-73-14, ¶3.2