1 fatality

10 Oct 2021: ROBINSON HELICOPTER COMPANY R44 II (N637HP) — Cornwall, NY

Cornwall, NY, United States

On 10 Oct 2021, a ROBINSON HELICOPTER COMPANY R44 II (registration N637HP) was involved in an aviation accident near Cornwall, NY. One person was killed. Investigators recorded the probable cause as: The non-instrument-rated pilot’s continued flight into deteriorating weather conditions, which resulted in a loss of control due to spatial disorientation. This summary draws on records from NTSB; 12 related events involving the same aircraft type or operator are linked below.

SourcesNTSBPrimary reportUpdated 1778583330Data APIEditorial standards

A Robinson R44 II helicopter was destroyed after crashing in Cornwall, New York, on October 10, 2021. The pilot was fatally injured. The helicopter departed from Floyd Bennet Memorial Airport and was en route to MacArthur Airport when the accident occurred.

History of Flight

On October 10, 2021, at 1357 eastern daylight time, a Robinson R44 II, registration N637HP, was destroyed when it crashed in Cornwall, New York. The pilot was fatally injured. The flight was conducted under Title 14 Code of Federal Regulations Part 91 as a personal flight.

According to fueling records, the pilot purchased 24.6 gallons of fuel before departing from Floyd Bennet Memorial Airport (GFL), Glenn Falls, New York, at about 1247. The intended destination was MacArthur Airport (ISP), Ronkonkoma, New York, where the helicopter was based.

ADS-B tracking data and National Weather Service (NWS) records indicated that as the helicopter flew south along the Hudson River at an altitude of 1,800 to 1,900 ft above mean sea level (msl), the overcast cloud ceiling height was decreasing. As it passed Poughkeepsie, NY (POU) at 1349, the reported ceiling was about 2,364 ft msl. At 1353, about 7 miles further south, the helicopter passed Stewart International Airport (SWF), where the ceiling was reported to be about 1,991 ft msl. At 1356, the helicopter began to climb.

NWS weather surveillance radar data showed that near the time the climb began, the helicopter encountered an area of light intensity echoes moving north-northwest.

At 1356:03, while at an altitude of 1,875 ft msl and a ground speed of 101 knots, the helicopter began to climb straight ahead. About 9 seconds later, it reached 2,025 ft, above the cloud ceiling reported at SWF (1,991 ft). About 5 seconds later, it began the first of two right turns. The first turn lasted about 9 seconds with an average turn rate of 7 to 11° per second.

From the start of the climb until about the end of the first right turn, the rate of climb continually increased, and airspeed continually decreased. At the end of the first turn, the rate of climb plateaued at about 2,400 feet per minute, and airspeed had decreased to about 45 knots. As the rate of climb plateaued, airspeed trend reversed and began to increase. The helicopter continued to climb for about 10 seconds as airspeed increased to about 95 knots. During this time, the second right turn began, lasting about 9 seconds at an average turn rate of 9-12° per second. At 1356:34, just before the end of the second turn, the helicopter began to descend rapidly from its peak altitude of 2,650 ft. About 6 seconds later, the descent rate reached 7,200 feet per minute. As it descended through 1,900 ft (just below the cloud ceiling at SWF), the descent rate reached 16,200 feet per minute. The tracking data ended about 2 seconds later, about 0.6 nautical miles north of the accident site. The entire flight was about 70 minutes, and the accident site was about 120 nautical miles south of GFL.

A witness on a hiking trail about 0.1 mile southeast of the accident site reported hearing the helicopter’s engine “falter and die.” One to two seconds later, he heard a swishing sound, which he attributed to the helicopter “spinning” or being “out of control”; however, he could not see the helicopter due to the tree canopy. Shortly thereafter, he heard a series of loud noises described as “backfires, at machine gun speed” for about 2 seconds, which he surmised was “the engine catching and the rotors slapping,” and then the sounds stopped. He reported visibility as “clear,” with clouds higher than mountain peaks, and no fog over the river.

A second witness located about 1/2-mile northwest of the accident site reported hearing a “loud noise” believed to be the helicopter’s engine. When he turned, he observed the helicopter in a nose-down attitude of nearly 90°, traveling at “high speed…straight down.” The witness lost sight as it descended below a tree line; shortly thereafter, he “heard it crash.” The engine noise was continuous until impact.

Personnel Information

The pilot held a private pilot certificate with a rotorcraft-helicopter rating. He did not have an instrument rating. His logbook showed no recorded actual or simulated instrument flight experience, and no instrument training was noted in the remarks section of any log entries. In November 2020, one flight was remarked with “IFR Conditions, asked for special VFR.” In May 2021, one flight was remarked with “low clouds, rain, autopilot through clouds.”

Aircraft Information

The helicopter was equipped with inflatable landing gear floats and a flight control stability augmentation system.

Meteorological Information

Weather conditions reported at the nearest reporting station (SWF) at the time of the accident included wind from 070° at 5 knots, visibility 5 miles in mist, ceiling overcast at 1,500 ft agl (1,991 ft msl), temperature 16° C, dew point 14° C, altimeter setting 30.20 inches of mercury. Data from other reporting stations indicated that overcast ceiling heights were decreasing from north to south. Infrared satellite imagery depicted a thick layer of clouds obscuring the accident site, with cloud tops near 27,000 ft. Weather surveillance radar imagery showed a small area of light echoes near the end of the ADS-B data, and a large area of light to moderate echoes between the accident site and the intended destination (ISP). The radar’s lowest scanning beam was centered about 5,800 ft msl over the accident site.

The Terminal Aerodrome Forecast (TAF) issued at 0723 expected marginal visual flight rules conditions at SWF, with visibility of 6 statute miles or greater and overcast cloud ceilings at 1,200 ft agl. The next closest TAF, also issued at 0723, expected instrument flight rules (IFR) conditions at Westchester County Airport (HPN), with visibility 5 statute miles in light rain showers and cloud ceiling overcast at 800 ft agl.

A national weather service AIRMET Sierra, current at the time of the accident, advised of IFR conditions due to ceilings below 1,000 ft agl and/or visibility below 3 miles in precipitation and mist. The AIRMET extended over the accident site and the planned destination. No other SIGMETs, convective SIGMETs, or CWAs were current for the area during the period.

A search of the FAA contract Automated Flight Service Station provider Leidos indicated that they, and no other 3rd party vendors using the Lockheed Flight Service System, had any contact with the pilot on the day of the accident. A search of ForeFlight records indicated the pilot had an account and had used the application at 1113, creating a route string for a flight from ISP to GFL via the Hudson River and back to ISP at 1,000 ft. Before the flight, the following airports were viewed in the application (including latest METAR, TAFs, and NOTAMs): Brookhaven Airport (HGV), Shirley, New York, and Minute Man Air Field (6B6), Stow, Massachusetts. The pilot did not view any weather imagery inside the application. No formal route briefing was requested, and there was no confirmation the pilot reviewed or searched for current inflight weather advisories. The pilot had multiple tablet devices in the cockpit, all too damaged to examine after the accident.

Wreckage and Impact Information

Examination of the accident scene revealed a debris path about 110 ft long, oriented on a heading of about 168° magnetic. It began with damaged treetops and broken limbs, where the tail rotor assembly and fragments of the tail rotor guard were located. The tail rotor blades were largely intact with leading edge gouge damage. The main impact crater, containing fragmented landing gear components, was about 60 ft from the damaged tree, with the main wreckage about 60 ft beyond the crater. The fuselage was fragmented and compressed from the nose toward the rear seats.

Flight control continuity could not be confirmed due to impact damage, and some system portions were not located. However, all flight control attachment fittings remained attached at their ends. Most flight control push-pull tubes were fractured, some in multiple locations, consistent with overload.

The overriding clutch operated normally and smoothly when rotated by hand. The upper drive sheave had impact marks on the front and rear faces adjacent to the clutch centering strut and fuselage frame tubes, respectively. A 3-inch imprint consistent with the starter ring gear was present in the grooves of the upper sheave. The main and auxiliary fuel tanks were impact damaged; their bladders were breached, and a trace amount of fuel remained. Some vegetation surrounding the impact crater and main wreckage showed evidence of fuel blight. A portion of each main rotor blade (about 4 to 6 ft) remained attached to the hub. Most of the spar of one blade also remained attached. Fragments of damaged main rotor blades were found along the debris path; the outboard 3 ft of one blade was found 273 ft northwest of the main wreckage. Some fragments had wood debris in the leading edge, others had chordwise streaks of brown residue consistent with tree material. Score marks oriented in the direction of rotation were found on a fuselage frame tube adjacent to the tail rotor drive shaft intermediate coupling. The filament of the low rotor speed warning lamp on the cockpit instrument panel was found intact and demonstrated hot filament stretching.

The engine crankshaft rotated smoothly when turned by hand at the cooling fan. Crankshaft and valvetrain continuity were confirmed as the crankshaft was rotated. Two cooling fan blades had impact marks/dents inline with airframe components located directly above the blades. Thumb suction and compression were attained on all six cylinders. A borescope examination of all cylinders revealed no internal damage or anomalies. Both magnetos were found separated from the engine. The left magneto was fractured in half, lower half not recovered. The right magneto’s upper case and capacitor were impact damaged but otherwise intact; it would not produce spark on any leads as found, but after replacing the damaged capacitor, it operated normally. The fuel servo was separated from the engine and partially fragmented. The mixture control arm was bent inward toward the servo and found in the idle-cutoff position. The push-pull mixture cable was not attached to the control arm, and attachment hardware was not present. The cable sheath was found stretched (in tension) in several locations.

Medical and Pathological Information

The Office of the Medical Examiner, Orange County, New York, performed an autopsy on the pilot. The cause of death was blunt impact injuries. Toxicology testing detected Tamsulosin in liver and muscle. Tamsulosin (Flomax) is an alpha blocker used to treat prostate hypertrophy and is acceptable for FAA medical certification.

Contributing factors

PilotAbility to respond/compensateEngine (reciprocating)