N301MASocata TBM 700 · 2017

On August 17, 2008, about 1455 Pacific daylight time, a Robinson R-22 Beta II helicopter, N301MA, impacted the terrain about two miles northwest of Olalla, Washington. The private pilot, who was the sole occupant of the helicopter, was killed in the accident. The helicopter, which was owned and operated by the pilot, was destroyed by the impact. The 14 Code of Federal Regulations Part 91 personal pleasure flight, which departed Bremerton National Airport, Bremerton, Washington, about 20 minutes prior to the accident, was being operated in visual meteorological conditions. No flight plan had been filed.

According to friends of the pilot, he recently created a landing pad area at his personal residence, and on the day of the accident he said he was going to fly from Bremerton Airport to his residence so he could land the helicopter on that pad for the first time. He and another individual drove to the airport together, and this second individual ultimately drove back to the residence after the pilot departed the airport in the helicopter.

According to the individual who drove to the airport with him, the pilot mentioned that it had been about a year since he had flown the helicopter, so in addition to adding fuel to it, he was going to spend some extra time looking it over and making sure it was ready to fly. Reportedly, after arriving at his hangar, the pilot spent between 60 and 90 minutes pre-flighting the helicopter prior to starting the engine. During that time, he took two five-gallon fuel containers to the fuel pumps, filled them with 100 low-lead aviation fuel, and added the contents of one of those containers to the fuel already in the helicopter.

While the pilot was preparing the helicopter for flight, the individual who drove to the airport with him asked if they could go along during the flight, but the pilot reportedly responded that since it had been so long since he had flown the helicopter, he wanted to take it up by himself. He also stated that he was going to do some practice maneuvers prior to departing the airport, and that he would not land back at his residence until the person who accompanied him to the airport had returned there in the car.

After starting the helicopter, the pilot lifted off and performed a hover-taxi to an open grassy area within the airport boundary. Once there, he spent about 10 minutes performing what appeared to be practice maneuvers, and then flew back by the person who drove to the airport with him. As he flew by, he waved, which was the pre-agreed upon signal that everything was alright, that he was going to head toward his residence, and that the individual on the ground should go ahead and drive the car back to the residence. Soon thereafter, the helicopter was seen departing the airport to the east, with no sign of anything out of the ordinary taking place.

Although the exact timing could not be determined, the pilot is believed to have arrived in the area near his residence about seven to ten minutes after departing the airport. According to a friend of the pilot, who lived directly across the street from the pilot's residence, once the pilot arrived in the area, he flew by the residence at least four different times. The witness said that each time the pilot flew by he would wave, and then depart the area for a few minutes prior to arriving for anther pass. According to this individual, during each pass the helicopter was about 200 feet above ground level (agl), and it was obvious that the pilot was not trying to land. The witness further stated that the helicopter's engine sounded like it was running smoothly during each of the fly-bys, and it was his opinion that the pilot continued flying around the area because he was waiting for the individual with the car to arrive back at the residence before he landed. During the last fly-by the helicopter was heading from north to south, and shortly after that pass, the helicopter descended to an impact with the terrain in the backyard of another residence located about one-half mile south of the pilot’s house.

Witnesses in the area of the impact had varying descriptions of what the helicopter sounded like just prior to and during its descent to impact. A couple of witnesses said that it sounded like a group of motorcycles were coming down the road together. One said that it sounded like it was making a loud grinding or "gear ticking" sound. One said that it sounded like the engine was running smooth, but at a very high rpm. Another said that they heard the engine sputtering or misfiring, and then turned to see the helicopter "fishtailing" as it started to descend. All of the witnesses said that from the time they first noticed the sound of the helicopter's engine, until they heard the sound of the impact, was no more than 10 seconds.

The 68 year old pilot held a private pilot certificate, with airplane single engine land(ASEL), airplane single engine sea (ASES), and rotorcraft-helicopter ratings. His last airmen's medical examination, a third class, was completed on July 3, 2008. On his medical application, the pilot indicated that he had a total of 2,201 hours total flying time, with 29 hours in the last six months. The pilot kept a separate logbook for his helicopter flying time, and the last entry in the log was on August 12, 2006, at which time he had 248.6 hours total helicopter time.

At 1455, the surface aviation weather surface observation (METAR) at Bremerton National Airport, which is located about 10 miles west-northwest of the accident site, indicated winds from 240 degrees at 3 knots, 10 statute miles visibility, clear skies, a temperature of 28 degrees Celsius, a dew point of 16 Celsius, and an altimeter setting of 29.76 inches of mercury.

The 1453 METAR issued at Tacoma Narrows Airport, which is located about 10 statute miles south of the accident site, indicated calm winds, a visibility of 10 statute miles, clear skies, a temperature of 27 degrees Celsius, a dew point of 16 degrees Celsius, and an altimeter setting of 29.74 inches of mercury.

The 1453 METAR for Seattle-Tacoma International Airport, which is located about 10 statute miles northeast of the accident site, indicated winds from 280 degrees at 06 knots, 10 statute miles visibility, scattered clouds at 11,000 feet, broken clouds at 19,000 feet, a temperature of 27 degrees Celsius, a dew point of 17 degrees Celsius, and an altimeter setting of 29.75 inches of mercury.

The helicopter impacted the terrain in the southwest corner of the backyard of a residence located on the west side of the 12600 block of Banner Road Southeast in Olalla. During the last part of its descent, its main rotor blades came in contact with the limbs of a tree at a height of about 20 feet above the ground. Immediately thereafter the airframe of the helicopter impacted a wood fence. The helicopter came to rest on its right side, partially inverted, about 10 feet west of the fence, in a dense growth of berry vines.

The forward part of the helicopter's belly had been compressed up and aft. The heaviest impact damage was on the left side of the belly. The skid cross tubes were bowed up at their ends, with the aft tube being disconnected from the left skid, and the forward tube being disconnected from the right skid. The toe of the left skid had separated at the forward strut. The SCAT tubing associated with the carburetor heat system had been crushed upward to a nearly flat condition, and the induction air filter box had been forced upward against the body of the carburetor with sufficient force as to result in the inflow end of the venture puncturing the bottom of the filter box. The carburetor heat cable housing was securely attached, but pinched tightly in the airframe structure by the upward crushing force. The carburetor heat slide was in the half-on position. The mixture control arm had been pushed almost directly upward, resulting in the bending of the mixture shaft. The mixture control arm had come to rest near the full lean position, with the distance between the cable clamp and the mixture cable retaining hardware being 0.7 of an inch, and the distance between the cable clamp and the end of the cable housing being 0.2 of an inch. The cable wire itself was found resting alongside (instead of through) the control arm attach bolt. It was lightly wedged between the bolt's lower washer and the end of the mixture control arm through which the bolt passes.

The first individuals to get to the scene reported that there was a significant amount of fuel pouring from the helicopter, and an impact hole was found on the inboard wall of the auxiliary fuel tank. Both fuel tank screens and the gascolator screen were found to be uncontaminated, and there was no evidence of water in the fuel samples. A very thin layer of orange discoloration settled from the fuel sample taken from the gascolator, but all other samples, including that from the carburetor, contained no separating discoloration. There was no evidence of pre-impact contamination of the air filter.

The lower left side of the cooling air induction fan lower scroll half contained an area of upward impact crushing, and the lower half of the scroll had four forward-to-aft fractures spaced about evenly across its circumference. The tail boom had separated around the first riveted seam aft of the tail boom attach fitting, and there was no indication of main rotor blade impact of the separated boom section. About a six inch section of the bottom of the boom, just forward of the second riveted seam, was crushed directly upward through about one-half of the boom structure, and the top half of the aforementioned riveted seam had pulled apart. The tail rotor drive shaft failed just forward of this second seam, and there was no evidence of torsional twisting at the failure location. Aft of the second seam, the tail boom, to include the horizontal and vertical stabilizers, was undamaged. The tail rotor gearbox was undamaged, and its temperature indicator tags (Teletemp) read between 140 to 150 degrees Fahrenheit. The tail rotor pitch change links were undamaged, and the pitch of both blades changed symmetrically when the tail rotor pitch control push/pull tube was moved forward and aft. The tail rotor blades both remained attached to their hub, and each showed a small amount of chord-wise staining streaks from contact with organic material near their leading edge. Neither showed leading edge rotational impact damage nor chord wise scarring of their outward facing surfaces. One blade was bent outward about 20 degrees about mid-span, and the other contained two small impact indentations, one being about four inches from its tip, and the other being about six inches from its root.

Both main rotor blades were bowed upward along the outboard 90 percent of their span, and each had between 12 to 15 chordwise compression wrinkles at intervals across their top surface. Neither blade displayed leading edge rotational impact indentations or chordwise scarring. The blue blade was torn through in a chordwise direction, except for its leading edge, about two feet from its tip. The outboard two feet beyond the tear was bent downward past 90 degrees. The rotor hub assembly, pitch link assembly, and swashplate assembly were all intact and determined to be functional once the cyclic push/pull control tubes, whose movement was restricted by the crushing of the of the bottom of the airframe, were cut where they exited the top of the fuselage.

Both rotor system drive belts had been pulled off of the driving and driven pulleys, but neither belt showed galling or melting of its inside surface. There was no rotational scaring of the driving pulley or of the cooling fan scroll. The forward flexplate was distorted in a forward and aft direction, and one of its ears was torn through at its attach bolt hole. The middle flexplate was distorted in a forward and aft direction, but all four ears remained intact and attached by their respective bolts. The aft flexplate was undamaged. None of the three flexplates showed torsional or twisting damage. The sprag clutch engaged/disengaged correctly and the main rotor gearbox was free to rotate.

The interior components of the cabin had been compressed forward and down, resulting in impact damage to the lower part of the center console. The left anti-torque pedal was found in the full forward position, and the throttle was in the full open position. The magneto switch was on BOTH. The clutch switch was in the engaged position, and the clutch guard was down (guarded). The carburetor heat knob was in the full in (OFF) position, and was not locked. The fuel selector was in the ON position, the primer was in and locked, and the mixture control knob was fully in, although its top edge was about 0.10 of an inch from the panel as a result of being bent downward. The cyclic friction, collective frictions were off, and cyclic trim knob was in the off/down position.

The King County Coroner's Office performed an autopsy on the pilot, and determined that the cause of death was, "Multiple blunt force injuries to the chest." The manner of death was determined to be accidental.

The Federal Aviation Administration's (FAA's) Forensic Toxicology Research Team performed a toxicology examination on specimens taken from the pilot, and that examination was negative for carbon monoxide and cyanide in the blood and negative for ethanol in the vitreous fluid. The examination was positive for the following drugs:

Carvedilol detected in the blood

Carvedilol detected in the urine

Diphenhydramine (0.088 ug/ml, ug/g) detected in the blood

Diphenhydramine detected in the urine

Omeprazole detected in the blood

The pilot’s most recent application for Airman Medical Certificate noted the use of carvedilol for high blood pressure and a history of seasonal allergies.

As part of the investigation, the engine was removed from the airframe and shipped in NTSB custody to Robinson Helicopter Company, Torrance, California. On October 8 and 9, 2008, the engine underwent an NTSB-monitored inspection and operational check on the Robinson dynamometer test stand. Once all impact damaged components where replaced with serviceable components, the engine was started and ran smooth at idle. Once the engine reached running temperature, the RPM was raised to 75 percent, and a magneto check was performed. During the check both magnetos operated within the single-magneto RPM drop specification. The engine was then run at various power settings, up to 2,700 RPM, for a total of twenty-five minutes. During the test run, the engine followed the acceleration and deceleration inputs of the throttle, and performed throughout the RPM range with no anomalies. At the end of the test run, the RPM was set at 2,100 RPM, and the mixture knob was slowly pulled toward a lean mixture until the engine stopped running.

After the test run, the oil filter was opened, and the filter screen was inspected. No significant contamination was present. The carburetor was then removed and its carburetor bowl was removed. There was no contamination inside the float bowl, the floats did not show any signs of leakage, and the accelerator pump was soft and pliable.

During the initial post-recovery field inspection of the engine, it was noted that the carburetor mixture control cable inner wire was no longer clamped through the eye in the A462-4 bolt that connected the wire to the mixture control arm. Instead, the wire was resting alongside the bolt, wedged between the bolt's lower washer and the end of the mixture control arm through which the bolt passes (see photo 12). It was noted that the torque-stripe paint between the retaining nut that held the bolt to the mixture control arm, and the palnut that locked the retaining nut in place was still intact. Gentle finger pressure was applied to the wire to see if it was secure in location it was found, but it readily slid from that position.

In order to establish whether the mixture control wire had been placed in its post-accident position prior to the accident sequence, the end of the mixture control wire and the attachment hardware (bolt, spacer, two washers, retaining nut) were sent to the NTSB materials laboratory, along with an exemplar mixture control arm, for further examination. That examination revealed that the marks/indentations across the center of the lower washer were consistent with the wire having been threaded through the eye of the bolt, and having been tightened in place in that position. There were no marks on the washers or on the side of the bolt that would have indicated the wire had been tightened in place in that position. There were scratches on the end of the control wire that were consistent with the wire having movement inward or outward while in place in the hole in the attach bolt, but it could not be determined if the scratches were the result of the wire being pulled from the eye of the bolt upon impact or during previous maintenance activity.

It was also noted that the mixture control knob shaft, which connects the knob to the wire, and slides in and out of the center console in which it is mounted, was bent about 30 degrees at the point on the shaft where the shaft would exit the panel when the knob is in the full rich position.

A review of the carburetor icing probability chart obtained from the DOT/FAA/CT-82/44 Publication (attached), revealed that the pilot was operating in an ambient environment where the accumulation of carburetor ice could be expected to occur at both glide and cruise power.

The Robinson R-22 Beta II is configured with a carburetor heat assist device. This device is designed to adjust the amount of carburetor heat applied based upon movement of the collective. Lowering the collective increases the amount of carburetor heat from the level preselected by the pilot, and raising the collective reduces the amount of carburetor heat from the level preselected by the pilot. Although the carburetor heat knob can be locked in the off position, rendering the carburetor heat assist device inoperative, according to Robinson Helicopter, in most ambient environments, and specifically that of the accident flight, the carburetor heat knob would be expected to be unlocked during takeoff. Prior to takeoff, with the rotor at an RPM of 100 percent, the pilot would select an amount of carburetor heat application that would place the indicator needle above the upper limit of the yellow arc of the Carburetor Air Temperature (CAT) gauge. Then during takeoff, with the carburetor heat knob unlocked, the carburetor heat assist device would reduce the amount of carburetor heat from its preselected position as the collective was raised, but when the carburetor heat knob reached the full in (OFF) position, the assist device could make no further reduction in carburetor heat. Then as the collective was lowered during the transition to cruise flight, the assist device would automatically apply an intermediate amount of carburetor heat. According to Robinson Helicopter, that application would normally place the carburetor heat knob somewhere in the range of about one inch out from the console. From that point on, it would still continue to be the responsibility of the pilot to monitor the CAT gauge and to select an amount of carburetor heat that would keep the CAT gauge needle out of the yellow arc. Although the carburetor assist device would then adjust the amount of carburetor heat applied based on the amount and direction of collective movement, it would not automatically keep the carburetor air temperature out of the yellow arc. Therefore, during maneuvering flight, the pilot must constantly monitor the CAT gauge, and make any necessary adjustments to the carburetor heat position.