NTSB Factual Report

HISTORY OF FLIGHT

On April 27, 1997, approximately 1245 Pacific daylight time, a Smith Aerostar 601P, N30LL, impacted the surface of Lake of the Woods about 23 miles northwest of Klamath Falls International Airport (Kingsley Field), Oregon. The private pilot, who was the sole occupant, received fatal injuries, and the aircraft sustained substantial damage. The pilot, who departed Bellingham International Airport, Bellingham, Washington, about three hours and twenty minutes earlier, was reportedly conducting a 14 CFR Part 91 flight en route to Midland, Texas, in order to visit a relative. The pilot, who had not filed a flight plan, was operating the aircraft in visual meteorological conditions at the time of the accident.

On the morning of the accident, the pilot went to his aircraft and contacted Aviation Northwest with a request for fuel. When the fueler arrived at the aircraft, the pilot first had him top off the fuselage tank. Then according to the fueler, the pilot pushed the fuselage tank drain valve in order to drain and inspect fuel from the bottom of the tank. When he pushed in the valve, it stuck in the open position, and he had to wiggle it in order to get it to pop out and close. Once it popped out, the fuel stopped flowing and there did not appear to be any leaks. After inspecting the fuel that had been drained from the fuselage tank, the pilot told the fueler to fill both wing tanks to the top because "...he needed every drop he could get." When fueling was completed, the aircraft had taken on 85.9 gallons of aviation fuel, and all tanks had been filled to the top.

About 15 minutes after the aircraft was refueled, the pilot started both engines and began a series of engine run-ups which lasted approximately 35 to 40 minutes. According to a witness, the aircraft remained in its usual parking spot while the pilot ran the engines at various power settings for varying lengths of time. Although one or both engines were occasionally at idle, it appeared to the witness that most of the time was spent with at least one of the engines, and sometimes both, at a high power setting. According to the witness, who said he had observed the run-up from a distance, he did not remember the engines running rough or anything unusual about their operation. At 0912, the pilot terminated the engine runs, and called for a taxi clearance. He was given clearance to runway 16, whereupon he taxied to the runway 16 run-up area and once again performed high-power run-ups on both engines for what was estimated to be about five to six minutes. At 0921, the pilot called Bellingham Tower for takeoff clearance and said that he would be making an eastbound departure. At 0923, the pilot departed Bellingham, made a left turn after takeoff, and was last seen climbing to the east. He did not contact Seattle Approach or Seattle Center during departure, and there is no record of the pilot making contact with any other FAA facility until he contacted Klamath Falls Air Traffic Control Tower at 1225. According to the FAA, the pilot had not filed a flight plan and did not receive an FAA-provided weather briefing prior to departure.

At 1225, the pilot of N30LL contacted Klamath Falls (Kingsley) Tower and said that he was inbound for landing, and estimated that he was 30 miles north-northeast of the airport. The tower told the pilot to report entering a right downwind for runway 32, and then gave him the current altimeter setting. The pilot advised the tower that he had Papa (Automatic Terminal Information Service Information Papa). There was no further contact between the aircraft and the tower until 1235, when the pilot transmitted "Klamath Tower Aero, ah this is ah Aerostar three zero lima, ah we've got ah fuel problem, we're a little low. I am over a lake and I don't know where the city is. I'm headed straight at a mountain with snow on top. Can you help me?" The controller asked the pilot whether he was transponder equipped, and the pilot responded that he was and that he was "...squawking one two zero zero." Then while coordinating with Seattle Center Sector 10 for a discrete transponder code for the aircraft, the controller asked the pilot if the lake he was over was a big lake. The pilot responded with "Yes, a great big lake with a highway on the north end of it." Believing that the pilot was over Klamath Lake (which was later confirmed by recorded time-annotated radar data), the controller told the pilot "Okay, you're over Klamath Lake, which is just northeast of the city. Ah, you can proceed southeast along the ridgeline and ah you'll come right into the city, and then right on ah the southeast side of the city you'll see the airport." The controller further advised the pilot that if he followed the freeway "...it will come right into the city." In addition the controller advised him that once he got over the city of Klamath Falls he would be able to see the airport. The pilot responded to these instructions with "Understand, zero lima lima." About 75 seconds after the pilot had said he was having a fuel problem, the controller told him to squawk a code of 3561, and the pilot repeated the squawk to the controller. The controller asked the pilot what his altitude was, and the pilot advised him he was descending through 7,000 feet. The controller then asked the pilot to confirm that he was low on fuel, but that the engines were still running. The pilot confirmed that was correct, and asked the controller to tell him again what the assigned squawk was. The controller repeated the squawk, and then, because Seattle Center had advised him that the aircraft was continuing to the west and had offered to provide vectors for the pilot, told the pilot he was headed westbound and advised him to contact Seattle Center on 127.6. Soon thereafter, the pilot made radio contact with Seattle Center and was instructed to fly heading one zero zero degrees. At 1239, about one minute after making contact with center, all radar and direct radio contact was lost with the aircraft. About three minutes after losing direct radio contact, the pilot of another aircraft (N116WC) reported he thought he had heard the pilot of N30LL reporting that he was heading toward Klamath Falls, but that one engine had lost power. About one minute after that report, the pilot of a third aircraft (N816JW) reported that he had heard the pilot of N30LL loud and clear, and that the pilot had reported he had lost both engines. There was no further direct or indirect radio contact with the pilot of N30LL.

The next reported sighting of the aircraft occurred at 1242, when it was spotted by individuals who were fishing on Lake of the Woods, which is located about 28 miles northwest of Klamath Falls. According to those individuals, when the aircraft was first observed it had come in low over the lake and had leveled off at about 200 to 300 feet above the surface. As it continued over the lake, which is approximately two and one-half miles long and three-quarter to one mile wide, it began to slow, and its nose began to rise. It reportedly continued to slow, and then suddenly rolled quickly to one side and went "...almost straight down into the lake." Although there were conflicting stories about whether the aircraft's engines were running or not, one pilot-rated witness was adamant that there were absolutely no engine sounds except one very loud engine roar of about two seconds duration that occurred just prior to the aircraft impacting the water. This individual said that he was not looking in the direction where the aircraft went down when it came in over the lake, and that he did not know there was an aircraft in the area until he heard a loud engine roar. Upon hearing the roar, he turned in the direction of the noise just in time to see a plume of water rise from where the aircraft had impacted the lake. He reported that he immediately went to the area where he had seen the plume and found numerous small pieces of debris. He also reported that there was no fuel slick and only a slight smell of aviation gas. He said that although there was no fuel slick, there was a slight oil slick, the source of which led him to the location of one of the engines in about 30 feet of water.

METEOROLOGICAL INFORMATION

The 1155 Aviation Routine Weather Report (METAR) for Klamath Falls reported winds 330 degrees at four knots, 30 statue miles visibility, few clouds at 3,000 feet, temperature 11 degrees Celsius, dew point minus three degrees Celsius, with an altimeter setting of 30.07 inches of Mercury. The Klamath Falls METAR for 1255, about 12 minutes after the accident, was winds 290 degrees at three knots, 30 miles visibility, few clouds at 3,000 feet, few clouds at 25,000 feet, temperature 13 degrees, dew point minus one degree, altimeter 30.06. Also, the pilot of another aircraft flying in the area of the accident at 1241, reported that it was "...severe VFR out there."

WRECKAGE AND IMPACT INFORMATION

The aircraft impacted the surface of the Lake of the Woods about 200 feet west of the Sunset Beach boat ramp, and sank in about 30 to 40 feet of water. The aircraft was recovered from the lake about a day and a half after the accident, and according to local environmental officials, during the time the aircraft remained submerged in the lake, only a very light sheen of fuel was seen on the water's surface. The wreckage was recovered from the lake in two major sections. The largest section included the fuselage from just forward of the instrument panel to just aft of the trailing edge of the wing flaps. This section included the inboard two-thirds of both wings, along the right engine, which was still partially attached to its mount. The second major section was made up of the aft portion of the fuselage and the empennage. Recovered separately were the first four feet of the nosecone, the outboard one-third of both wings, the left engine, which had separated from its mount, and various smaller pieces of structure and system components. After recovery, the wreckage was taken to Independence, Oregon for further examination.

Examination of the wreckage revealed that most of the cabin, to include the instrument panel and center console, had been crushed, twisted, and torn apart by the force of the impact. Much of the fuselage skin and the leading edge of both wings had been distorted aft by hydraulic impact forces. All three gear were in the retracted position, and the aft fuselage/empennage was intact and relatively undamaged. Both propellers remained attached to their respective engines, and the outboard three-fourths of all three blades of the left propeller were bent aft approximately 60 degrees. There were no chord-wise scratches on any of these blades, but one blade had two small leading edge indentations. All three blades on the right propeller were found aligned approximately 90 degrees to their plain of rotation (feathered), and two of the three blades were nearly straight. One blade was bent back about 20 degrees along the outer one-half of its span, and the last three inches of the tip were bent further back another 30 degrees. Numerous random scratches to the front face of all three of these blades occurred during the recovery process, but there was no consistently clear pattern of chord-wise scratches on any of them. The throttle for the right engine was found retarded to idle, but the throttle for the left engine was found in the full forward (maximum power) position. Except for a fracture in the top compression ring in the number two cylinder of the right engine, the inspection of the engines and turbocharges revealed no inconsistencies or anomalies except those that appeared to have resulted from the forces of the impact. In addition, no contamination was found in the engine oil filters or the fuel injector servo inlet fuel screens. On both engines, mechanical continuity was established from the crankshaft to the pistons, and from the crankshaft to the valves. Both crankshafts were free to be rotated by hand, and compression was established in all cylinders of both engines (except for the number one cylinder of the left engine, which had two bent pushrods).

Because the pilot had reported that he had a "fuel problem," the aircraft's fuel system was examined in detail, with the following findings:

1. Both fuel selector knob mechanisms were destroyed, and the knobs were free to rotate in the panel.

2. The fuel tank caps on the left wing and fuselage fuel tanks were in place and secure. The section of the right wing skin that contained the fuel tank filler neck was not recovered.

3. Both wing tanks had been ruptured, but the fuselage tank was intact.

4. The outlet screen in both wing tanks were in place and were free of obstructions.

5. The metal fuel lines in the wings and fuselage were intact with no sign of leakage (except for the lines between the fuel flow sender and the engines, which had been severed during the impact).

6. The short rubber hoses used to connect portions of the fuel lines in the wings and wing roots were intact and showed no signs of leakage. Although the outer (red) layer of these two-layer hose sections had a number of small surface cracks in them at the point where the metal hose clamps had been tightened over them, none of the cracks penetrated the thicker inner (black) layer of the hose.

7. The L-shaped fuel line connectors in the wheel wells were intact and showed no sign of leakage.

8. The reinforced rubber hose connectors between the end of the fuel line and the right and left fuel sumps were intact and showed no sign of leakage.

9. All four fuel shutoff valves on the fuel sump assembly moved freely and could be opened and closed by hand.

10. The left, right, and fuselage fuel sump assembly moved freely and could be opened and closed by hand.

11. All three fuel sump drains were intact, worked correctly, and showed no sign of leakage.

12. Fuel shutoff valves number one and number four were in the closed position, and valves number two and three were in the open position (double crossfeed setup).

13. Both auxiliary fuel pumps were attached to a 28 volt battery for testing, and both motored and produced a suction on their intake side.

14. Air was blown through both fuel-flow senders, and neither appeared to have any obstruction or restriction to flow.

There were no anomalies or discrepancies existing within the fuel system except those that appeared to have resulted from the impact forces and/or the recovery operation.

Because the right throttle was found at idle and the left throttle in the full-forward position, the procedures delineated in the Engine Failure/Restart checklist in the owner's manual were reviewed. These procedures call for the throttle of the failed engine (both engines in this case) to be placed in the "Cracked 1/2 inch open" position.

MEDICAL AND PATHOLOGICAL INFORMATION

A forensic toxicology examination was completed on the pilot by the Federal Aviation Administration's Toxicology and Accident Research Laboratory. Although there was no carbon monoxide or cyanide found in his blood, and no ethanol detected in his vitreous fluid, the drug screening portion of the examination detected the following:

0.945 (ug/ml, ug/g) Nordiazepam detected in blood. 0.068 (ug/ml, ug/g) Oxazepam detected in blood. 0.722 (ug/ml, ug/g) Nordiazepam detected in kidney fluid. 0.070 (ug/ml, ug/g) Oxazepam detected in kidney fluid. 0.876 (ug/ml, ug/g) Chlordiazepoxide detected in blood. 0.346 (ug/ml, ug/g) Norchlordiazepoxide detected in blood. 1.480 (ug/ml, ug/g) Chlodiazepoxide detected in kidney. 0.658 (ug/ml, ug/g) Norchlordiazepoxide detected in kidney. 0.130 (ug/ml, ug/g) Diphenhydramine detected in blood. 0.372 (ug/ml, ug/g) Diphenhydramine detected in kidney. Propranolol was detected in blood. Propranolol was detected in kidney. 4.200 (ug/ml, ug/g) Acetaminophen detected in blood.

These results indicate the presence in the pilot's blood of chlordiazepoxide and three of its active metabolites, norchlordiazepoxide, nordiazepam, and oxazepam. Chlordiazepoxide (Librium) is a tranquilizer often used to treat anxiety and tension. At sufficient levels it can ha

NTSB Probable Cause

The pilot's failure to set the throttle of his second failed engine to 'Cracked-1/2-Open' as called for in the Engine Failure/Restart checklist, followed by a high-power engine surge. Factors include the pilot's delay in landing for refueling, the pilot becoming lost/disoriented, drug impairment, and fuel exhaustion.