LAX97FA046

During IFR arrival, the pilot was cleared for an ILS runway 7 approach after being vectored to intercept the localizer from the south side. Last assigned altitude was 2000' msl. Radar data showed the aircraft (acft) continued across the localizer toward rising terrain. Air traffic controllers (ATC) issued instructions to turn back to the localizer, but the pilot did not follow. ATC then issued urgent instructions to turn & climb. (Later, they said he calmly acknowledged their transmissions, but continued toward rising terrain.) Radar contact was lost, but the airplane finally turned south. However, it hit avacado trees on the west side of sloping terrain at about 440' msl, while in a steep bank. No preimpact mechanical anomaly was found that would have resulted in the accident. The pilot's former wife recalled instances when he had miss-set the autopilot OBS; on those occasions, he had similarly deviated from the localizer at intercept & resisted her suggestion to take over manually & fly the approach, but persisted in trying to make the autopilot fly the approach. A flight instructor, who administered a biennial flight review 1 month earlier, found the pilot's instrument flight skills weak & refused to certify his instrument competency. Toxicology tests of tissue from the pilot indicated the presence of a narcotic pain reliever and its metabolite. Tests of kidney tissue showed 0.337 mcg/ml Propoxyphene & 2.7 mcg/ml Norpropoxyphene; tests of muscle tissue showed 0.385 mcg/ml Propoxyphene and 1.9 mcg/ml Norpropoxyphene.

History of flight

On November 20, 1996, at 1108 hours Pacific standard time, a Beech 95-B55, N210WW, impacted mountainous terrain at Goleta, California, while being radar vectored for an instrument approach to Santa Barbara, California. The aircraft was destroyed and the private pilot was fatally injured. Instrument meteorological conditions prevailed and the aircraft was operating on an instrument flight plan. The aircraft departed Long Beach, California, at 1024.

The pilot contacted the Hawthorne (California) Automated Flight Service Station at 0806 on the morning of the accident and requested a briefing for a flight from Long Beach to Santa Barbara. The pilot then filed two IFR flight plans: one for the flight to Santa Barbara and the second for the return flight to Long Beach later in the day.

The aircraft departed Long Beach at 1024 and the pilot proceeded via a series of radar vectors until checking in with Santa Barbara TRACON, east radar, at 1052 in the vicinity of KWANG intersection, 22 miles east of Santa Barbara at 4,000 feet. After passing KWANG, the pilot was given radar vectors to the ILS runway 7 final approach course at Santa Barbara. The aircraft first flew a downwind leg on a westbound heading south of the airport, during which the aircraft was instructed to maintain an airspeed of at least 160 knots. At 1100, when the aircraft was approximately 5 miles south of the airport still on a westbound heading, the pilot was given a frequency change to the Santa Barbara TRACON west radar frequency. After checking in with the west radar controller, the aircraft was issued a descent to 2,000 feet at 1102, and then, at 1104, was issued a right turn to a heading of 350 degrees, which placed the aircraft on base leg.

At 1106, when the aircraft was 3 miles from NAPPS intersection, the pilot was instructed to turn right to heading 060 and maintain 2,000 feet until established on the localizer and was cleared for the ILS Runway 7 approach. Eleven seconds later, the pilot asked "zero six zero or three six zero." Five seconds later, the controller radioed that the pilot was flying through the localizer and issued instructions to continue the right turn to a 100 degree heading for the localizer intercept. The pilot acknowledged the clearance 8 seconds later.

At 1107:12, the controller instructed the pilot to turn left to heading 180 degrees, but on learning he was in a right turn, affirmed that he wanted the pilot to turn to heading 180. At 1107:35, the controller transmitted "I don't know what he's doing, Baron zero whiskey whiskey cancel approach clearance fly heading one eight zero" and at 1107:41, the controller's training supervisor transmitted "climb, climb." At 1107:42, the controller transmitted "appear to be northbound, climb and maintain four thousand, turn right now heading one eight zero." At 1107:48, the pilot replied "right now at one eight zero." This was the last communication from the pilot. At 1107:56, the controller transmitted "Baron zero whiskey whiskey, you're heading toward the hillside, you need to crank that aircraft around now heading one eight zero please" and at 1108:10, the controller transmitted "Baron zero whiskey whiskey, climb and maintain four thousand." There were no further communications from the pilot.

The controllers told the Safety Board that the pilot's voice remained calm and unhurried despite the urgency of their calls. It appeared to the controllers that the target on their radar initially turned toward the localizer and then veered off to the north. The Santa Barbara TRACON utilizes ARTS II radar equipment and there was no recorded radar data. The wreckage was distributed on a southerly course.

Witnesses living nearby reported that low overcast clouds obscured the mountains at the time of the accident. A pilot, who was inside his house, reported that it sounded to him like the aircraft was about 200 feet above his house (at 50 -75 feet msl), and yet his wife, who was outside, could not see the aircraft in the clouds. The witnesses reported that the engines sounded normal, perhaps at greater than approach power. One witness thought the aircraft may have been trying to climb and another reported that the engines sounded abnormal briefly before the accident.

According to the pilot's former wife, the pilot was under a great deal of stress from their recently completed divorce. She described the pilot as a very strong-willed person who aggressively fought the divorce. She described the divorce as "brutal" and believed that he was very distraught, so much so that his children had discouraged him from flying.

The pilot's former wife is also an instrument and multiengine rated private pilot, and has often flown with the pilot in the accident aircraft since they were both licensed in 1976. They last flew together in late 1995. Regarding the pilot not intercepting the localizer after the vector to final approach prior to the accident, she recalled four or five instances when she had been with him and he miss-set the OBS on the HSI causing the autopilot to turn the aircraft away from the final approach course instead of intercepting it. It was his practice, when being radar vectored for approach, to use the autopilot in heading mode by slewing the heading bug on the HSI. On the four or five instances in question, he had not set the OBS to the final approach course heading, and, using the heading bug, he turned to intercept the localizer and engaged approach mode on the autopilot. The autopilot then flew the heading on the heading bug until the localizer was intercepted, at which point the autopilot switched to localizer track mode and turned to whatever heading was set on the OBS. She emphasized that the turn was surprisingly steep. She added that in some of these instances she had suggested to him that he disconnect the autopilot and hand fly the approach until he got the aircraft back on course, but that he resisted and persisted in trying to make the autopilot fly the approach.

She speculated that, in the accident, he may have had the OBS set to a northerly course which turned the aircraft back to the north after the localizer intercept and by the time he took control of the aircraft from the autopilot, together with the urgency of the controllers instructions, the aircraft was out of control.

The pilot's son spoke with his father by telephone the night before the accident. His son felt that although his father was upset by the divorce, he was coping adequately and was emotionally fit to fly.

Personnel information

The pilot's most recent logbook was not located after the accident. His family was able to produce a logbook in which the last entry was dated 12/9/94. At that time the logbook showed the pilot's total flying time to be 1,959 hours, of which 447 hours were in multiengine aircraft. Totals for instrument flight time had not been carried forward; however, there were entries showing that the pilot had flown 1.6 hours of actual instrument time and 1 hour of simulated instrument time in the previous 6 months. On his medical certificate application dated July 29, 1996, the pilot reported 3,500 hours total pilot time with 10 hours in the last 6 months. On his previous medical certificate application, dated July 25, 1994, he reported 3,700 hours total pilot time with 70 hours in the last 6 months. On an insurance application dated October 12, 1996, the pilot reported total flying time of 2,000 hours, 692 hours in the accident aircraft type, 40 hours in the last 12 months, and 20 hours in the last 6 months.

A pilot logbook found in the wreckage contained only one entry for a biennial flight review on October 11, 1996. The flight instructor who performed the BFR told the Safety Board that the pilot had contacted him and was very hurried to obtain a BFR and an instrument competency check. The flight instructor reported that th