United States of America

During the preflight briefing, the pilot was informed of reduced visibility and low ceiling in the vicinity of the destination airport, at the approximate time of the planned arrival. Upon arriving in the Atlanta area, the pilot was issued radar vectors to a final for the localizer runway 31 approach. The pilot was also given the current Atlanta altimeter setting, and was cleared for the localizer runway 31 approach. The airplane collided with a 60-foot tall light pole at a nearby baseball complex 2 miles short of the runway. The weather observation from the Hartsfield International Airport indicated that visual weather conditions prevailed at the time of the accident. However, according to a witness at the accident site, the weather conditions were foggy with reduced visibility. The wreckage distribution path was 2,467 feet right of the localizer course. The minimum descent altitude for this approach, using Atlanta's altimeter setting, was 1260 feet. The ground check of the localizer and DME facility was within normal operating range. Examination of the aircraft navigational radios also tested within normal ranges. The average field elevation in the vicinity of the accident site is 800 feet. The pilot's toxicological examinations detected pseudoephedrine (decongestant), phenylpropanolamine (decongestant), and chlorpheniramine (antihistamine). No samples were available to quantify the blood levels of these medications.

A Mitsubishi MU-2 departed Salt Lake City, Utah, and climbed to 16,000 feet MS on an IFR flight to Pocatello, Idaho. While in cruise flight, the MU-2 encountered structural icing conditions. According to radar data, the MU-2 began slowing from a cruise speed of about 190 knots with slight deviations from heading and altitude. The airspeed decreased to about 100 knots, and the flight crew declared an unspecified emergency, then radio contact was lost. The MU-2 began a right turn, then it entered a steep descent and crashed. The pilot of a Beech 1900 (about 12 minutes in trail of the MU-2), stated that he encountered moderate rime icing at 16,000 feet. The Beech pilot activated his deice boots (3 times) and descended to 12,000 feet to exit the icing conditions. The MU-2 flight manual warned that during flight in icing conditions, stall warning devices may not be accurate and should not be relied upon; and to minimize ice accumulation, maintain a minimum cruise speed of 180 knots or exit the icing conditions. An investigation determined that the captain of the MU-2 was aware of deficiencies in the timer for the deice boots, as well as other maintenance deficiencies. The captain's medical certificate was dated 11/17/94; he was providing executive transportation for compensation under an agreement for "contractual flights," under 14 CFR 91. Although icing conditions were forecast in the destination area, no icing was forecast for the en route portion of the flight.

The aircraft was destroyed after an uncontrolled descent into the Pacific ocean about 14 miles west-southwest of El Segundo, California. The pilot was presumed to have been fatally injured. According to a company search pilot, visual meteorological conditions prevailed at the accident area about 1.5 hours after the time of the accident. No flight plan was filed for the positioning flight which originated at Oxnard, California, on the morning of the accident for a flight to San Diego, California. The aircraft departed Oxnard on a special VFR clearance. The tops of the clouds were reported to be about 1,200 feet msl. The aircraft transitioned southbound through the NAWS Point Mugu airspace. The Point Mugu radar approach control monitored the aircraft on radar for about 25 miles. The pilot was subsequently given a frequency change to SOCAL Tracon. There was no contact made with that facility. A search was initiated when the aircraft failed to arrive at the intended destination. A review of the recorded radar data revealed the aircraft was level at 1,800 feet msl and then climbed to about 2,000 feet msl, at which time it disappeared from radar.

The pilot received abbreviated weather briefing for emergency medical service (EMS)/air ambulance flight. Before flight, he expressed anxiety about possible low visibility for landing and timely transport of dying patient. During ILS runway 03 approach (rwy 03 approach), the aircraft remained well above the glide slope until close to the middle marker; aircraft's speed decreased from 153 to 100 kts, while vertical speed increased from 711 feet/min to about 1,250 feet/min descent. About 1 mile from runway and 500 feet agl (in fog), the aircraft abruptly turned left of localizer course and gradually descended with no distress call from pilot. The aircraft hit a pole, then flew into a building and burned. Low ceiling, fog and dark night conditions prevailed. Pilot (recent ex military helicopter pilot) had logged/reported 3,500 hours of flight time and about 150 hours in multiengine airplanes, but there was evidence he lacked experience with actual instrument approaches in fixed wing aircraft; he had difficulty with instrument flying during recent training and FAA check flights. No preimpact mechanical problem was found with aircraft/engines. No ILS anomalies were found. Flight nurse was using cellular phone, but no evidence was found of interference with aircraft's navigational system. Visibility and ceiling at destination were less than forecast at time of pilot's preflight weather briefing. Paramedic was only survivor.

Before the ferry flight, the pilot (PIC) & inexperienced copilot noted the left & right, float-type, underwing, fuel gauges indicated about 3,900 & 4,050 lbs of fuel, respectively. After takeoff, they noted that the cockpit gauges showed an opposite fuel imbalance of 4,100 & 3,600 lbs in the left & right tanks. Due to this indication, the PIC crossfed fuel from the left tank to both engines for about 30 min to rectify the perceived fuel imbalance. Later as they approached the destination, the left tank was exhausted of fuel, & the left engine lost power, although the left gauge indicated about 500 lbs of fuel remaining in that tank. The PIC then crossfed fuel from the right tank to both engines, & left engine power was restored. ATC vectored the flight for an emergency ILS runway 3 approach. The PIC was distracted during the approach & maneuvered the airplane to re-intercept the localizer. About 500' agl in IMC, both engines lost power. During a forced landing at night, the airplane struck a raised berm & was damaged. No evidence of fuel was found in the left tank; 125 gal of fuel was found in the right tank. Unusable fuel was published as 3 gal. During an exam of the engines & fuel system components, no preimpact failure was found. Historical data from the manufacturer indicated that when the airplane had a low fuel state, unporting of fuel tank outlets could occur during certain maneuvers. This information was not in the Convair 340 flight manual, although unporting of the outlets on this flight was not verified.

The pilots obtained a complete weather briefing for their IFR flight. Before landing at their destination, they attempted to obtain an airport advisory on Unicom, but received no response. They landed on the runway in low visibility with the runway lights visible. After touchdown, the left main landing gear contacted a berm or snowbank on the left side of the plowed area, and the pilots were unable to maintain directional control. The aircraft drifted off the left side of the runway and came to rest on a reverse heading. The runway, which was 100 feet in width, had been plowed to about 45.5 feet width along the centerline. No notams had been filed concerning the partially plowed condition of the runway. The second officer (commercial pilot) noted that during the weather briefing, the pilot-in-command had been advised of thin, loose snow on the runway.

The flight was cleared for the VOR/DME approach to runway 17 at the Marco Island Airport. The CVR recorded conversation between the pilot and co-pilot reference to the approach, specifically the MDA both in mean sea level and absolute altitude for a straight-in-approach to runway 17. The flight crew announced that the flight was landing on runway 35. The flight crew did not discuss the missed approach procedure nor the circling minimums. The flight continued and the co-pilot announced that the flight was 5 miles from the airport to descend to the MDA to visually acquire the airport. While descending about 8.5 feet of the left wing of the airplane was severed by a guy wire about 587 feet above ground level from an antenna that was 3.36 nautical miles from the threshold of runway 17. The tower is listed on the approach chart that was provided to the flight crew. The airplane then rolled left wing low, recovered to wings level, then was observed to roll to the left, pitch nose down, and impacted the ground. A fireball was then observed by witnesses. The altimeters, air data computer, and pilot's airspeed indicator were last calibrated about 8 months before the accident. The co-pilots altimeter was found set .01 high from the last known altimeter setting provided to the flight crew. The CVR did not record any conversation pertaining to failure or malfunction of either the pilot or copilot's HSI, the DME or Altimeters. There were no alarms from the VOR/DME monitoring equipment the day of the accident. The flight crew of another airplane executed the same approach about 30 minutes before the accident and they reported no discrepancies with the approach. The MDA for the segment of the approach between where the tower is located is no lower than 974 feet above ground level.

The airplane was circling to land on runway 22 after executing a VOR/DME approach. The airplane impacted the ground approximately one quarter mile northeast of the runway 22 threshold. The wreckage path covered a distance of approximately 350 feet. Control continuity was established. Airframe, engine and navaid examination revealed no abnormalities. The left wing and horizontal stabilizer leading edges had approximately one-eighth inch of rime ice adhering to their leading edges. Two witnesses reported seeing the airplane rolling from the left to the right. The Eagle River AWOS was not available on a VHF radio frequency, due to radio frequency congestion at the O'Hare International Airport, Chicago, Illinois.

The aircraft impacted mountainous terrain in controlled flight during hours of darkness and marginal VFR conditions. The flight was being vectored for an instrument approach during the pilot's 14 CFR Part 135 instrument competency check flight. The flight was instructed by approach control to maintain VFR conditions, and was assigned a heading and altitude to fly which caused the aircraft to fly into another airspace sector below the minimum vectoring altitude (MVA). FAA Order 7110.65, Section 5-6-1, requires that if a VFR aircraft is assigned both a heading and altitude simultaneously, the altitude must be at or above the MVA. The controller did not issue a safety alert, and in an interview, said he was not concerned when the flight approached an area of higher minimum vectoring altitudes (MVA's) because the flight was VFR and 'pilots fly VFR below the MVA every day.' At the time of the accident, the controller was working six arrival sectors and experienced a surge of arriving aircraft. The approach control facility supervisor was monitoring the controller and did not detect and correct the vector below the MVA.

At 1350 cst, a McAlester FSS specialist gave a preflight briefing to a Cessna 421 pilot concerning IMC (instrument meteorological conditions) along the route & advised that VFR flight was not recommended. Cloud tops were at 12,000', & freezing level was at 1,600'. A PIREP at 1416 cst reported light mixed icing from 6,400' to 9,000' at Oklahoma City. At 1424 cst, the pilot departed Tulsa (VFR), then radar service was terminated. No further communication was received from the airplane. Radar data showed that it climbed westerly, reaching 9,800' at 1440 cst; during the next 88 seconds, its heading & altitude deviated until it descended through 3,200'. Ground witnesses saw the airplane descend out of low clouds in a 'flat spin' & crash. No preimpact mechanical failure was found. The airplane's gross weight was about 150 lbs over its maximum limit. In November 1995, the pilot received 10 hrs of Cessna 421 simulator training; his instructor noted in training records that he met minimum standards for VFR, but 'under IMC conditions,' he 'could not maintain altitude within 1,300 feet or heading within 40 degrees.' Postmortem toxicology tests showed 0.079 mcg/ml Nordiazepam (metabolite of Valium) in kidney fluid, 0.044 mcg/ml Desipramine (metabolite of Imipramine, an antidepressant) in spleen fluid, 0.733 mcg/ml Diphenhydramine (Benadryl) in spleen fluid, & 0.353 mcg/ml Diphenhydramine in lung fluid. These medications are not approved by the FAA for use while flying. The airplane was equipped for flight in icing & IFR
conditions.