Plane crash map Find crash sites, wreckage and more

N512HH accident description

Go to the Oregon map...
Go to the Oregon list...

Tail numberN512HH
Accident dateOctober 19, 1996
Aircraft typeRobinson R22 BETA
LocationHalsey, OR
Additional details: None

NTSB description


On October 19, 1996, approximately 0926 Pacific daylight time, a Robinson R22 Beta helicopter, N512HH, operated by Avia Flight Services Inc. of Corvallis, Oregon, crashed in a grass seed field approximately 1 mile southwest of Halsey, Oregon. The helicopter was destroyed and the recently certificated private pilot, a 21-year-old Brazilian national who was the helicopter's sole occupant, was killed. The 14 CFR 91 flight had departed Corvallis with an unknown destination. Visual meteorological conditions prevailed and no flight plan had been filed.

The pilot was a student in Avia's commercial helicopter pilot training program. According to the chief executive officer (CEO) of Avia Flight Services, the accident flight was an unscheduled solo flight and not a training syllabus ride. Avia's CEO stated that the pilot had not been scheduled to fly until noon that day, when a dual instructional flight with an instructor was scheduled. He stated that the accident pilot came in to the office about 0800, determined that a helicopter was available, and signed the helicopter out to fly solo; the CEO reported that the pilot stated he would be out from 0800 to 1000 but did not provide an itinerary or get an instructor to dispatch the flight. Avia's CEO stated that this solo flight was unexpected, in light of intentions previously stated by the pilot to Avia to fly only dual flights until completion of the training program due to financial concerns. Avia's CEO reported that the pilot refueled the accident aircraft with 10 gallons of 100LL aviation gasoline before departure, and was also seen doing a preflight inspection of the aircraft. He further stated that the accident pilot departed Corvallis, a non-towered airport, between 0830 and 0900, without making a radio call (a procedure he described as non-standard.) The CEO reported that the crash occurred outside Avia's normal practice area northeast of Corvallis (the accident site was approximately 11 nautical miles southeast of the Corvallis airport), and stated that the school's flights did not normally operate in the Halsey area on weekends due to noise abatement rules (the accident occurred on a Saturday.)

No witnesses to the accident were identified. Local fire/rescue and law enforcement authorities responded to the crash scene after being notified of the crash by a local resident via a 911 emergency call. The pilot was determined to be dead at the scene.


According to Avia's CEO, the pilot had arrived at the flight training school in February 1996. He held a private pilot certificate with airplane single-engine land and rotorcraft-helicopter privileges, with a date of issuance of October 11, 1996, eight days before the accident (his logbooks indicated that he passed his private pilot airplane checkride in June 1996, and his helicopter checkride in October 1996). Avia personnel reported to responding law enforcement officials that the pilot was due to complete his commercial pilot training within the next two weeks, requiring approximately nine more hours of dual flight instruction to complete the program.

Avia personnel described the pilot as a highly motivated and dedicated student who loved flying and had good piloting skills, but reported that they had noted a change in his behavior and attitude during the two weeks prior to the accident. According to statements made by the Avia CEO and the pilot's roommate to investigators and local law enforcement personnel, the pilot had been "very depressed" over an impending breakup with his girlfriend in Brazil for approximately the past two weeks. As an example of this change in attitude and behavior, the pilot's flight instructor reported to law enforcement officials that he had scolded the pilot the day before the accident for being 40 minutes late to a flight lesson. Avia's CEO also told law enforcement officials that the pilot had previously displayed very strong emotions, and that he had cautioned the pilot not to fly with other things on his mind. The pilot's roommate told law enforcement officials that "[the pilot] had been crying at night and on 10/18/96 [the day before the accident] [the pilot] had taken all of the letters and card[s] he had gotten from his girlfriend and he had thrown them in the trash."

The Avia CEO reported a number of perceived unusual behaviors by the pilot immediately before the accident flight, including leaving a favorite flight helmet (which had been given to him by his father, and which several Avia personnel stated he always wore whenever he flew) in the office during the flight, failing to respond to a casual greeting from his flight instructor, and taking off from Corvallis without making a radio call.

Avia's CEO later reported that while going through the pilot's helmet bag in the Avia office after the accident, he discovered photocopies of a newspaper article about a recent fatal Oregon helicopter accident (SEA97FA001, Boeing Vertol BV-107-II, N196CH, Canby, Oregon, October 4, 1996) folded together with a stack of faxed letters from the pilot's girlfriend. The faxes, which were in Portuguese and undated, were provided to a Portuguese translator. The translated letters did not provide any clear indications as to the pilot's state of mind or intentions, but did indicate that the pilot and his girlfriend had argued (and possibly made up) on at least one occasion; that the pilot may have had difficult relations with the girlfriend's mother; and that the girlfriend was unhappy with the extended overseas separation from the pilot.

In a letter faxed to the NTSB IIC dated December 11, 1996, the pilot's father stated that, while the father was in Brazil at the time and therefore "not the right person to tell details about the behavior of my son in his last moments in Corvallis", he nevertheless believed that the pilot "wanted to be back home (he was homesick) in spite of some hints which might point to the possibility of non-accident. It is my opinion that this idea can not [sic] be taken seriously because no firm indication [exists], as far as I learnt during my staying in Corvallis [after the accident]." It was also noted in the sheriff's report that the pilot had plans to go shopping with his roommate in Portland that afternoon to buy gifts for his family in Brazil, and that the pilot was due to return home to Brazil soon and planned to discuss relationship matters with his girlfriend at that time. The pilot's roommate told a sheriff's detective that upon learning of the accident, he returned to their apartment to look for any notes which the pilot may have left, and found none.

The pilot's logbook, which was recovered from the aircraft wreckage, indicated that the pilot had a total of 48.7 hours of helicopter time, of which 26.4 hours was helicopter pilot-in-command. All of the pilot's helicopter time had been logged in the R22. The logbook contained an instructor endorsement indicating that he had received the type-specific training required by Special Federal Aviation Regulation (SFAR) 73 to act as pilot-in-command of R22 helicopters.


In its April 1996 Special Investigation Report (SIR), "Robinson Helicopter Company R22 Loss of Main Rotor Control Accidents" (NTSB/SIR-96/03), the NTSB reported that the R22 is highly responsive to flight control inputs. Additionally, according to the SIR, the R22 utilizes a low-inertia main rotor, which is subject to rapid RPM decay if collective pitch and/or engine power are improperly managed by the pilot. Data presented in the SIR indicated that the R22 had experienced a high proportion of fatal loss-of-control accidents. R22 fatal loss-of-control (LOC) accidents (as defined in the SIR) accounted for 37 per cent of total R22 fatal accidents between 1981 and 1994. The NTSB SIR also noted that according to one FAA study, approximately 80 per cent of the R22 flight hours logged in 1989 were for instructional flights.

In 1995-96, the FAA issued a series of Airworthiness Directives (ADs, specifically ADs 95-02-03, 95-04-14, 95-26-04, 95-11-09, and 96-11-08) and one SFAR (SFAR 73) aimed at preventing R22 LOC accidents. These measures included: imposing additional operating limitations on the helicopter; requiring installation of a placard in clear view of the pilot reading "LOW-G PUSHOVERS PROHIBITED"; instituting type-specific training and experience requirements for piloting and giving flight instruction in the R22; requiring installation or upgrade of a throttle/collective governor on all R22 helicopters; mandating an increase in the RPM at which the low RPM warning light and horn activate; and prohibiting flight with the governor off except in cases of malfunction or emergency procedures training. These additional measures were intended to minimize the possibility of pilot mismanagement of main rotor RPM leading to loss of control of the helicopter. The logbooks of the accident helicopter documented compliance with the above requirements, including installation of a Robinson Helicopter Company KI67-2 governor kit in accordance with AD 96-11-08 on June 28, 1996. Additionally, the pilot's logbook entries and endorsements indicated that he was in compliance with the type-specific R22 training and experience requirements specified by SFAR 73 (see PERSONNEL INFORMATION above.)

The FAA-approved R22 Pilot's Operating Handbook (POH) gives the helicopter's best rate-of-climb airspeed as 53 knots indicated airspeed (KIAS), its normal landing approach speed as 60 KIAS, and its best autorotation speed as 65 KIAS. Additionally, the 1995 FAA ADs added instructions to the POH urging pilots to maintain speeds above 60 KIAS during normal cruise, and to adjust speed to not lower than 60 KIAS in the event of an inadvertent encounter with moderate, severe, or extreme turbulence (the same ADs prohibited intentional flight of the helicopter in moderate, severe, or extreme turbulence.)

The R22 POH gives the R22's minimum power-on rotor RPM as 97%, and its minimum power-off rotor RPM as 90%. According to a copy of the R22 POH on file in the NTSB office in Seattle, Washington, incorporating revisions through April 14, 1989, the LOW RPM warning light and horn activate when main rotor speed drops below 95%. AD 96-11-08 mandates an increase in this threshold to between 96 and 97 per cent.

The height/velocity diagram in the R22 POH indicates that at an airspeed of 59 KIAS, flight below approximately 20 feet above ground level (AGL) should be avoided.

The POH gives the helicopter's never-exceed speed as 102 KIAS. The Robinson party representative stated to the NTSB IIC that in flight testing of the R22, the type did not exhibit characteristics of retreating blade stall below speeds of 130 to 140 KIAS, and that this speed could not be attained by the helicopter in level flight due to power limitations.


Temperature at Eugene, Oregon (16 nautical miles south of the accident site) at 0856 was reported as 5 degrees C, and dewpoint was 3 degrees C. According to an FAA carburetor ice prediction chart, this combination of temperature/dewpoint conditions is conducive to serious icing at cruise power. Avia's CEO told investigators that during the morning of the accident flight, some pilots at Corvallis reported that their airplanes were experiencing carburetor ice formation during taxi-out for takeoff. On the accident helicopter, the carburetor heat control was destroyed, the carburetor heat valve was fully open and the carburetor air temperature gauge needle displayed an "off scale high" indication.

The winds in the 0856 Eugene automated observation were reported as 160 degrees at 5 knots, and the sea level pressure was reported as 1,016.7 millibars (30.02 inches Hg.) Eugene is 365 feet above sea level.

The 0956 Eugene automated observation reported that rain began there at 0921, and was reported in the observation as light. The 0956 observation also indicated that between 0856 and 0956, sky conditions dropped from few clouds at 6,000 feet at 0856 to few clouds at 800 feet with a broken 3,000-foot ceiling at 0956. The 0956 observation listed the winds as 190 degrees at 9 knots, temperature as 6 degrees C, dewpoint as 4 degrees C, and sea level pressure as 1,017.6 millibars (30.05 inches Hg.)

The Corvallis automated observation for 0935 listed winds from 190 degrees at 5 knots, scattered clouds at 6,000 feet, temperature 5 degrees C, dewpoint 2 degrees C, and altimeter setting 30.03 inches Hg. At 0955, this had changed to a 6,000-foot broken ceiling with a reported temperature of 6 degrees C.

According to the county sheriff's report, an Avia helicopter pilot who flew out from Corvallis to the accident scene with the CEO reported that he noted no adverse weather conditions in that area at the time that would have adversely affected the flight.


The helicopter wreckage was examined at the accident site on October 19-20, 1996. The wreckage was located in a level grass seed field and was distributed in a generally east-to-west pattern approximately 250 feet long, running roughly parallel to and just to the south of an east-west road. A north-south power line was located approximately 1/4 mile east of the wreckage site, but the power line exhibited no evidence of a strike from the helicopter. Likewise, no evidence of a wire strike was noted on the helicopter wreckage.

The easternmost point of the wreckage area was a large ground scar approximately 1 foot deep. Two generally east-west oriented ground scars, each approximately the length and width of one of the helicopter's main rotor blades, extended to the west of this ground scar, and the helicopter's two landing skids were speared into the ground on the north edge of this ground scar at approximately a 66-degree angle to the ground (with the exposed ends pointing approximately east in azimuth.) Both skids had penetrated approximately 14 inches into the ground. The helicopter's empennage and the outer two inches of one of the tail rotor blades were also located in this large ground scar. A burn area approximately 66 feet long and 60 feet wide extended immediately to the west of the ground scar.

The aft 4 feet of the helicopter's tail boom (less empennage), with the tail rotor (less the outermost two inches of one blade) in place, was located about 75 feet west of the large ground scar. The aft tail boom section was separated from the forward section along a rivet line. Approximately 4 additional feet of the tail rotor drive shaft and tail rotor control push-pull tube protruded forward out of the forward end of the separated tail boom section. The forward end of the tail rotor drive shaft exhibited a torsional overstress fracture signature. The tail rotor could be turned by rotating the tail rotor drive shaft by hand, and the tail rotor blade pitch could also be changed by actuating the tail rotor control push/pull tube by hand. No evidence of contact between the main rotor and tail boom was observed.

A second smaller ground scar was found approximately 125 feet west of the large ground scar. The main wreckage of the helicopter, consisting of the fuselage and forward tail boom section with main rotor blades, engine, and transmission still in place, was located approximately 25 feet beyond the smaller ground scar (150 feet west of the large ground scar.) The main wreckage had come to rest on its right side headed approximately north. The two main rotor blades were still attached to the main wreckage, were continuous from tip to tip, and were both bent or broken upward. Both swash plate arms and one main rotor pitch link exhibited overload breaks, but the main rotor hub remained securely attached to both blades and to the mast, and no evidence of main rotor hub-to-mast contact was observed. There was also no evidence observed of main rotor blade contact with the fuselage. Within the main wreckage, no evidence of preimpact flight control system or rotor drive train discontinuities or malfunctions was found. Within the engine, the oil pump casing was cracked, and the engine-driven oil pump would not turn by hand. Disassembly of the oil pump revealed no evidence of intern

(c) 2009-2011 Lee C. Baker. For informational purposes only.