Lessons from a Learjet Crash
Aviation Safety's Brian Jacobson takes a critical look at the June 1994 crash of a Mexican-registered Learjet 25D at Washington's Dulles airport, killing both pilots and all ten passengers. The brand new Learjet captain — who did poorly during his recent transition training at FlightSafety International — inexplicably attempted two ILS approaches when Dulles tower was reporting RVR far below minimums. He missed the first approach, and his luck ran out on the second try.
Pilots transitioning from one type of aircraft to another have a responsibility to themselves and to their passengers to ensure that they are comfortable and safe in the new aircraft. Others, from instructors to designated check airmen to company managers, have the job of overseeing the process and ensuring that when a recently upgraded pilot makes his first flights in a new type of aircraft, there is no compromise in safety. Sometimes, a breakdown occurs in the upgrade process that allows a pilot to "step up" before he or she is ready to do so. Such a breakdown appears to have been involved in the 1994 crash of a Learjet 25D at Dulles airport.
The airplane, operated by TAESA, a Mexican airline and charter operator, left Mexico City at 11 p.m. on June 18, 1994 with 10 passengers and two pilots. The destination was Washington, D.C. with a fuel stop planned in New Orleans. After landing at Lakefront Airport at 0125, there was a delay because the Customs agent was waiting for the aircraft at New Orleans International, on the other side of the city. The airplane didn't leave New Orleans until 0347.
Near the Dulles airport. the crew had to hold for four and a half minutes while ATC helped a Mooney pilot who had declared an emergency. The Learjet then was cleared direct to the Armel VOR at 11,000 feet, and Dulles Approach vectored the aircraft to Runway 1R for an ILS approach. The ATIS was reporting the weather as ''indefinite ceiling 600, sky obscured, visibility one-half fog, temperature 71, dew point 71, wind 140 at four."
The Lear 25 received its approach clearance at 0608, and the crew switched over to the tower frequency. Four minutes later, the crew declared a missed approach and advised that they'd like to try it again. United Flight 186 Heavy, a DC- 10, reported a missed approach shortly after the Learjet did. The captain of the United flight briefly discussed trying the ILS to l9L, because the runway visual range (RVR) on that side of the airport was higher, but he decided, instead, to proceed to his alternate airport. He may have felt that by the time he got around to the other approach, the visibility would be just as bad as the south side of the airport. The Learjet captain made no mention of trying the other approach.
As the controller vectored the Learjet back to the final approach course, the touchdown RVR for Runway 1R was only 600 feet and the rollout RVR was 4,000 feet. The airplane crashed about three-quarters of a mile from the runway, killing all aboard.
The 27-year-old captain had upgraded from copilot only two months before the accident and had flown about 87 hours as PIC in Learjets. His total flying time was only 1,706 hours, including 1,314 as copilot. His upgrade training had been performed at FlightSafety International. It included 14 hours of ground training and 12 hours of simulator time. His instructor had noted a defective instrument scan and poor use of the flight director. Notes written on the last day of training stated, "Pilot needs more CRM [cockpit resource management] training to be competent as PIC. Below standards for PIC. Additional training offered and declined."
The instructor told investigators that the pilot allowed the airplane to become airborne on every rejected takeoff on the last day of training and that his instrument approaches did not meet ATP standards. He said the pilot was interested in additional training but believed his company needed him to fly the line. Back home, the pilot was required to fly 10 hours as PIC with an instructor pilot aboard and to take a written test and a flight check before receiving his Learjet type rating.
Before sending the pilot for his check ride, TAESA's operations director requested a confidential evaluation from FlightSafety but received only a copy of the training record carried back by the pilot. He again requested a confidential evaluation, including instructor's notes, but was advised that the notes were only for internal use. He did, however, receive a letter stating, "During his simulator training, he demonstrated satisfactory flying skills when flying the aircraft under normal conditions. He requires emphasis in crew management and decision making skills during his training to upgrade to captain. [He] needs to improve his airmanship and command skills, especially when operating under the stress of abnormal and emergency situations. His most notable strength is his ability to smoothly fly the aircraft under normal operations. He displayed excellent qualities when acting in the capacity of first officer. [He] can be considered for upgrade to pilot-in-command. During upgrade training, situational awareness under high workload conditions should be emphasized. He should fly with a strong training captain or first officer during his upgrade."
It's obvious that the training facility thought the pilot would receive more training from his company before being upgraded to captain — and he did. He flew 10 hours with an instructor to demonstrate that he could act as PIC. But the pilot's weaknesses may not have been apparent if the 10 hours were spent flying in relatively good weather and low-workload situations. It's doubtful that the instructor would have simulated emergency situations during revenue flights.
The fact that the captain had so little overall experience should have caused the training facility and the company to increase their awareness of his progress. The company's operations director, who twice had requested complete information on the pilot, may have been satisfied by the statement that the "pilot can be considered for upgrade to pilot-in-command.' Had the instructor's notes been furnished, instead of the letter of evaluation, the company might have been more concerned about the ability of the pilot to upgrade so quickly to captain. NTSB recognized that the language the training facility used in its letter was "permissive in nature" and probably was taken as an approval of the applicant for upgrade to captain.
Several recent airline accidents have been attributed, in part, to the pairing of crew members who were relatively inexperienced in the aircraft they were flying. The copilot of the Learjet had a total of 852 flying hours, 426 in Learjets. TAESA is sure that the captain was flying the aircraft due to the weather conditions and the fact that the copilot was handling the radios.
On the first approach, the aircraft intercepted the localizer about 14 miles from the runway threshold, but the approach was never stabilized. The airplane wandered back and forth between the limits of the localizer course. The vertical path was erratic and finally wound up well above the full flydown limit of the glide slope. The airplane then descended at more than 2,000 fpm until it was back on the glide slope, but it was also at the full right limit of the localizer. The aircraft leveled at 600 feet MSL and maintained that altitude until it was about a mile north of the departure end of the runway. We can only wonder if the crew intentionally flew well past the missed approach point while trying to find the airport visually, despite the poor weather. It was only when the controller asked if they were going around that the copilot replied in the affirmative.
On the second approach, the localizer tracking was better, but the airplane descended at an average rate of 1,300 fpm between 1,300 feet and about 500 feet MSL. The decision height was 513 feet. NTSB noted that due to the resolution of the radar data, the aircraft could have descended as low as 350 feet. It then climbed at 1,300 fpm to 600 feet and, five seconds later, began a 3,000-fpm descent that lasted until it hit the ground.
It's obvious that the captain lacked situational awareness on both approaches. On the first approach, he flew to the right of the runway at 600 feet with no apparent consideration of a missed approach until queried by the tower. TAESA's operations manual states that a descent to the runway will not be attempted if the ceiling and visibility are below the approved minimums. Minimums for a Category 1 ILS are given as 200 feet and a half mile visibility (RVR 2,400), or RVR 1,800 if centerline and touchdown lights are in use. The touchdown RVR was 1,000 feet when the first approach was initiated and 600 when the second was begun.
Why did the captain initiate two approaches when the weather was below landing minimums? We can only speculate that he felt his alternate, Baltimore, had similar weather. But there is no record that he checked Baltimore's weather during the approach sequence. (NTSB did not include the actual BWI weather in its report, so we don't know if the alternate was open.
Fatigue may have played a part in this accident. The captain had been awake for 11.5 hours following a three hour nap he took in preparation for the flight. But what is more important is that both pilots flew all night — a disruption of their normal habits.
Chasing the Needles
Another thing to consider is the conduct of the approaches and how that relates to the instructor's notes on the captain's performance in the simulator. Instrument students attempting to perform a precision approach with an instrument scan that is too slow for what's happening around them often will "chase the needles," with large control deflections that result in oscillations to both sides of the horizontal and vertical centerlines. The student tends to concentrate on the needles instead of the approach. That means a loss of situational awareness as the pilot becomes so intent on getting the needles centered that the rest of the instruments are forgotten.
The training facility instructor's notes show that he was not happy with the Learjet pilot's performance during the first simulator session. His instrument scan was deficient, and he made poor use of the flight director. Although the pilot did better during the second session, he ultimately was deemed not up to ATP standards, which are used in assessing a pilot for a type rating.
A flight director can be a great aid during an instrument approach, but it is only an aid. The pilot must still scan the rest of his instruments to be certain that the flight director is providing proper guidance and to maintain an awareness of where the airplane is in relationship to the airport and the decision height or minimum descent altitude for the approach.
The training facility's instructor did not note what the pilot's problem was with the flight director. With 1,300 hours in the right seat of the Lear, he should have been thoroughly familiar with its operation, even if the airplane had only one instrument on the captain's side. But I have seen some pilots who simply don't understand the flight director and don't use it. It is apparent from the localizer and glide slope deviations that the captain was not using the flight director during the first approach. Unless he was totally "stressed" because of the weather, the earlier unexpected hold and the long night of flying, use of the flight director should have kept him more aligned.
It's possible that he did use it on the second approach, because his horizontal track was more in line with the extended runway centerline. His vertical tracking, however, was still not up to speed. Perhaps, as he got closer to the runway, he took his eyes off the instruments in an attempt to locate the runway — a job the copilot should have been doing. It's possible that when the airplane dropped below the glide slope to a dangerous elevation, the captain pulled the nose up to the command bars of the flight director, then lost control because he was, once again, looking for the runway. We can only speculate on what happened in the cockpit in the moments before the impact because the aircraft was not equipped with a cockpit voice recorder.
NTSB said that a ground proximity warning system (GPWS) might have prevented this accident. A warning would have been issued approximately 64 seconds before impact at an altitude of 1,200 feet MSL. Although the board recommended that all jets with six or more passenger seats be required to have GPWS, the FAA does not agree that the requirement be extended beyond turbojet aircraft with 10 or more passenger seats.
This Learjet was fitted with eight passenger seats, but there were 10 passengers aboard. Six of them were children, all five years of age or older, and it is likely that they were illegally sharing seats or sitting on the floor.
The sole responsibility for the safety of flight rests with the pilot-in-command. The captain of the Learjet knew from his encounter with the training agency's simulator that his approaches were not as good as they should have been. Although he did fly with an instructor and passed a check ride for his type rating, there is no evidence that his performance had improved.
He attempted a second approach after a sloppy first one. Even if the captain believed that by positioning himself better he might see the runway environment, a United flight had declared a missed approach after he did and proceeded to its alternate. That, and the low RVR readings, should have been a tipoff that he wouldn't get into Dulles until the weather improved. Perhaps, the pilot was concerned about his fuel. The report makes no mention of the airplane's fuel status. But if it was a factor, the pilot should have been keeping tabs on fuel, time and the weather; if it looked tight, he should have headed for a suitable alternate with plenty of time to spare.
Flight instructors should take note of the discrepancy between the training facility instructor's notes and the letter that was sent to TAESA saying the pilot could be considered for upgrade. The wording of any verbal or written report is crucial in similar circumstances. If you are not the person writing the reports on your students, ask to see them before they are sent. Be certain there is no room for misinterpretation of the facts by anyone who will review the report.
Whether a pilot is upgrading from the right to left seat of a Learjet or from a Cessna 172 to a Cessna 182RG, there is a standard that must be met before he or she is signed off. Pilots-in-training and flight instructors must not compromise safety of flight by flying or allowing others to fly new aircraft before they are competent. Just because you feel good in the "new" light single or twin in VFR conditions doesn't mean you can handle it when the weather is bad or workload is high.
There are too many things that can happen when you're in the clouds that will distract you and take your attention from the job at hand: flying the airplane. The Lear captain may have been suffering from fatigue and been distracted by the weather. He may have been searching for the airport instead of flying the airplane. Ultimately, this may have been the last link in the error chain that caused this crash.