N33PXCessna 560 · 2002

On September 1, 2009, about 1425 central daylight time, a Robinson R44 Raven I helicopter, N33PX, was substantially damaged when it struck trees, terrain and a house near Jackson/Hawkins Field (HKS), Jackson, Mississippi. The pilot in command (PIC) was seriously injured, and the second pilot was fatally injured. The combination currency flight for the PIC and familiarization flight for the second pilot was operated under the provisions of Title 14 Code of Federal Regulations Part 91. Visual meteorological conditions prevailed, and no flight plan was filed.

According to recorded air traffic control (ATC) voice and radar information, the helicopter first contacted Jackson Approach about 1404, when it was still on the ground at Flowood Industrial STOLport (MS90), Flowood, Mississippi. The approach controller instructed the helicopter to contact approach once it was airborne, and assigned it a discrete transponder code. Shortly thereafter, the helicopter reported that it was airborne and northbound. The helicopter target was acquired by the ATC radar facility located at Jackson-Evers Airport (JAN), Jackson, Mississippi. The first radar return associated with the helicopter appeared at 1405:55, when the helicopter was at a radar-indicated altitude of 300 feet above mean seal level (msl), and 2,400 feet north of MS90. Shortly after that, the helicopter contacted approach control again, advised that there was a change of plans, and requested to proceed to HKS for practice at "Area 1." "Area 1" was the designation for a helicopter practice area at HKS, which was located in the grassy area to the east of runway 16/34, and to the north of runway 11/29. The radar data ground track showed the helicopter turn from a northerly heading to a south-southwesterly heading, at an altitude of 700 feet msl.

When the helicopter was approximately 4 miles southeast of HKS, Jackson Approach informed the helicopter that radar services were terminated, and instructed it to contact HKS air traffic control tower (ATCT). The helicopter contacted HKS ATCT, requested approval for hover practice, and the request was approved by HKS ATCT. The radar ground track showed that the helicopter turned to the northwest and headed for the southeast corner of Area 1. As the helicopter descended towards HKS, it temporarily disappeared below Jackson Approach radar coverage; the last return acquired during this approach was at 1411:20, and an altitude of 400 feet. According to the ATCT local controller, the helicopter practiced hovering for a short time, and then landed in Area 1.

The helicopter then requested "right closed traffic" for Area 1, and was cleared "for the option" by the ATCT controller. The helicopter's radar return was re-acquired at 1420:22, at an altitude of 500 feet. The helicopter flew the downwind leg at an altitude of 900 feet msl, and a lateral offset of approximately 3,800 feet to the east of Area 1. The helicopter made another approach to Area 1, and the last radar return for this approach was acquired at 1422:53, with an indicated altitude of 400 feet msl. The local controller reported that the helicopter conducted a "touch and go," and began another circuit of the traffic pattern. The helicopter re-appeared on radar at 1424:08, at an altitude of 600 feet. Radar returns indicated that the helicopter made an immediate right turn for a second downwind leg that was offset approximately 1,700 feet east of Area 1. The local controller again cleared the helicopter for the option.

The local controller then communicated with an aircraft waiting to depart runway 34, and when she tried to visually locate the helicopter to ensure traffic separation, she did not see it. She made several calls to the helicopter, but it did not respond. The last three radar returns from the helicopter were at 1424:37.220, 1424:41.840 and 1424:46.550. The radar return altitudes were 1,000, 800, and 600 feet, respectively. The last recorded radar return was located 700 feet northwest of the accident site.

Two pilots from the Mississippi Department of Public Safety (DPS) were standing outside a hangar on the west side of HKS, and one of them saw the helicopter make an approach to Area 1. He then watched it taxi to the north, taxi back to the south, and then climb out to the north for a circuit of the traffic pattern. When the helicopter was on the downwind leg approximately due east of his position, the DPS pilot saw it begin "falling rapidly," and he pointed it out to the other DPS pilot. The other DPS pilot, who had flown Robinson helicopters, initially responded that the high descent rate was not unusual for Robinson helicopters, but as he watched it, he agreed that the descent rate did seem high. After the helicopter had descended a few hundred feet, the two DPS pilots observed it enter a right bank of about 15 degrees. Both DPS pilots stated that prior to the bank, the helicopter's pitch attitude was approximately level. Both pilots watched the helicopter disappear behind the treeline to the east. They realized that the helicopter had likely crashed, and in response, began preparing their helicopter for flight.

Another witness was walking down his driveway to meet the postman, when his attention was drawn to the helicopter by its noise, which he described as repeated "clup" sounds. The witness saw the helicopter descend rapidly towards him from the north, strike a tree, and enter "into a spin." It then struck a second tree and "fell into the house," disappearing from view behind the house across the street from the witness. The witness crossed the street and went around the right (east) side of the house to the helicopter, where he saw a person in the left front seat. The postman went around the west side of the house to the helicopter, and noticed that there was another person in the helicopter as well. Both witnesses used their mobile telephones to summon assistance.

Both occupants of the helicopter were inspectors from the Jackson, Mississippi, FAA Flight Standards District Office (FSDO). The PIC was seated in the right seat, and the second pilot was seated in the left seat for the flight.

Pilot in Command (PIC)

According to FAA information, as of the date of the accident, the PIC was employed by the FAA for about 3 years. He held multiple certificates and ratings, including commercial pilot and flight instructor certificates with rotorcraft-helicopter ratings. According to information compiled from the PIC's personal logbook and FAA records, as of the date of the accident, he had approximately 22,000 total hours of flight experience. This included approximately 221 total hours of helicopter experience, which included approximately 6 hours of R22 experience, and 81 hours of R44 experience. The records indicated that 27.9 of those hours were accrued without a check- or safety-pilot on board. The balance of his helicopter experience was primarily in Bell 206 and Schweizer 300 series helicopters. His most recent FAA second-class medical certificate was issued in April 2009.

The PIC's logbook indicated that his initial helicopter flight occurred in April, 2006, in an R44. His first recorded solo flight in a helicopter was conducted on November 7, 2007, in a Schweizer 300 helicopter. The PIC successfully completed his flight test for a commercial helicopter certificate on November 15, 2007. At that time he had 99 total hours of helicopter flight experience, including 39 hours of dual, and no solo, flight experience in Robinson helicopters. On January 22, 2008 the PIC successfully completed his flight test for a helicopter instructor rating. At that time he had 138 total hours of helicopter flight experience, including 60 hours of dual, and no solo, flight experience in Robinson helicopters. Both the commercial- and instructor- flight tests were conducted in a Schweizer 300 helicopter.

The most recent R44 entries in the PIC's logbook were dated February 5 and February 12, 2009, and both indicated that the flights were conducted in the accident helicopter. The February 5 flight was also the PIC's most recent flight review. According to the Jackson FSDO computerized records of the PIC's flight history, excluding the accident flight, his most recent flight experience in an R44 was on July 23, 2009, and his most recent flight experience in a helicopter was in a Bell 206, which occurred on July 30, 2009.

A logbook endorsement dated February 2007 stated that the PIC had completed the awareness training in accordance with paragraphs "(a)(3) under (b)(5)" of Section 2 of Special Federal Aviation Regulation (SFAR) No. 73. A logbook entry dated April 23, 2008 indicated that the PIC completed the Robinson Helicopter Company (RHC) Pilot Safety Course, and this was confirmed by a representative of RHC. The logbook also contained two endorsements dated February 5, 2009. One stated that the PIC met the experience, flight training, and demonstrated ability requirements to "provide instruction on the general subject areas of Section2(a)(3) and the flight training identified in Section 2(b)(5)(iii) in a Robinson R-44 helicopter." That endorsement did not explicitly specify that the paragraph references were from SFAR73. The endorsement was a component of the requirements to enable the PIC to provide flight instruction in the R44. The other logbook endorsement certified that the PIC had "satisfactorily completed the Flight Review required by FAR 61.56 and SFAR73 Section2(a)(3) and the flight training identified in Section 2(c)(2) and (3) in an R44." Both endorsements were signed by the FAA Safety Team (FAAST) team manager, who was a colleague at the Jackson FSDO.

Despite the logbook endorsements and FAA certificates which indicated that the PIC was qualified to provide flight instruction in Robinson helicopters, no evidence that the PIC ever provided flight instruction in a Robinson helicopter was able to be located.

Subsequent to the accident, the PIC could not recall any specifics regarding the conduct or events of the flight.

Second Pilot

According to information provided by the FAA, as of the date of the accident, the second pilot was employed by the FAA for about 1 month. The second pilot held multiple certificates and ratings, including airline transport pilot and flight instructor certificates, each with a rotorcraft-helicopter rating. His most recent FAA first-class medical certificate was issued in October 2008.

According to information that the second pilot provided on his employment application to the FAA, he had approximately 16,800 total hours of flight experience, including approximately 4,250 hours in helicopters. The second pilot's application stated that he had "over 3,000" hours of flight experience as pilot in command, including "1,000+" hours as flight instructor, and "500 to 1,000" hours as second in command, in US Army military helicopters. The cited helicopter type was the "BH 205," also referred to as the Bell UH-1H "Huey," which was a turbine-powered helicopter with a maximum gross weight of 10,000 pounds. His application also cited "1- 99" hours in "Bell Single Piston-powered" and "100 - 499" hours in "Hughes Single Piston-powered" helicopters. The investigation was unable to locate or review the second pilot's flight logbooks that documented his helicopter experience.

The second pilot's most recent helicopter piloting experience was in 2000. The accident flight was his first flight in a Robinson helicopter. No records of any Robinson-specific training or endorsements, including those required by SFAR73, were located. No definitive evidence regarding whether the second pilot did or did not manipulate the controls was discovered.

According to the manufacturer and the FAA, the helicopter was manufactured and registered to the owner in 2007. It was equipped with a Lycoming O-540 series piston engine. Power transfer from the engine to the rotor system was accomplished via four "vee-belts" that linked the engine (lower) and rotor system (upper) sheaves. The upper sheave was equipped with an overrunning clutch which permitted the sheave to rotate the main and tail rotor shafts, but prevented the reverse. An electric actuator maintained proper vee-belt tension by moving the upper sheave to change its distance from the lower sheave. The actuator was enabled via a clutch switch in the cockpit. The fuel system was gravity-flow. Fuel capacity was 30.6 gallons in the main tank, and 18.3 gallons in the auxiliary tank. The maximum certificated gross weight was 2,400 pounds.

The collective control was equipped with a "twist grip" throttle, and an electronic governor system was installed to maintain engine rpm by adjusting the throttle when the collective control was moved. The Pilots Operating Handbook (POH) stated that the governor was "easily overridden by the pilot," was only active above 80 per cent engine rpm, and could be switched on or off via a toggle switch on the end of the right-seat collective control. The POH stated that the governor was designed "to assist the pilot" in controlling engine rpm in the normal operating range, but that it "may not prevent over or under-speed conditions generated by aggressive flight maneuvers."

The helicopter was equipped with a single aural warning horn system; the system was designed so that the horn would sound whenever the main rotor rpm decreased to a value of 97 percent or less. The helicopter was also equipped with a low rotor rpm cockpit warning light designed to illuminate at the same rotor rpm threshold. The POH "Starting Engine and Run-Up" checklist in the "Normal Procedures" section required the pilot to verify the proper activation of the warning horn and light prior to every flight. The horn and the light each received their signal on a shared circuit from the Low RPM Warning Unit. The Warning Unit received a signal from one of the two Hall Effect transducers on the input (drive) yoke of the main rotor gearbox, and the other transducer provided a signal to the rotor tachometer.

According to the owner of the property-management company that the helicopter was registered to, the helicopter was primarily for his personal and business use, but was also rented to FAA inspectors from the Jackson FSDO for inspectors' currency and proficiency flights. No other persons flew the helicopter. The helicopter was based at MS90, and kept in the owner's hangar there.

According to the maintenance records, the most recent annual inspection was completed in January 2009. At that time, the airframe and engine had a total time in service of 198 hours. A pocket notebook that was used by FSDO personnel to record flight times was recovered from the helicopter. It indicated that the hour meter on the helicopter registered 236.8 hours at the start of the accident flight. At the accident site, the hour meter was observed to register 237.1 hours.

The 1353 weather observation at HKS included winds from 330 degrees at 8 knots, 10 miles visibility, few clouds at 4,700 feet, temperature 27 degrees C, dew point 14 degrees C, and an altimeter setting of 30.13 inches of mercury. The 1453 HKS observation reported winds from 320 degrees at 6 knots, 10 miles visibility, few clouds at 4,900 feet, temperature 28 degrees C, dew point 15 degrees C, and an altimeter setting of 30.11 inches of mercury.

Recorded ATC communications indicated that the helicopter was in radio communication with Jackson Approach Control over a period of 3 minutes and 35 seconds, and then with the HKS ATCT over a period of 13 minutes and 55 seconds. The helicopter was tracked by ATC radar, albeit discontinuously, for a total duration of 18 minutes and 51 seconds.

According to FAA and commercial information, the elevation of HKS was 270 feet msl. The non-federal ATCT was operated and staffed by a private company, Robinson Aviation, Inc (RVA, not affiliated with RHC). A letter of agreement (LOA) between several private and government entities, but not including the FAA, described the "procedures and responsibilities to promote safe and efficient control of helicopters operating at" HKS. The LOA included descriptions and graphics of Area 1, and specified that "helicopters shall fly a pattern altitude of 1,050 feet" msl.

The accident site was located approximately 1/2 mile east of HKS, in a residential neighborhood. Wreckage path information indicated that the helicopter first struck a tree slightly left of the flight path, approximately 60 feet above ground level (agl), and about 135 feet prior to its final resting location. It then struck a second tree to the right of its flight path, approximately 25 feet agl, and 55 feet prior to its final location. The first ground scars were located about 30 feet prior to the final location of the helicopter. The helicopter came to rest on the ground, lodged into a wooden porch at the back of the house. With the exception of the outboard segment of each main rotor blade, all components were located within about 15 feet of the main wreckage. The wreckage path was aligned on a magnetic course of approximately 150 degrees. There were no indications of pre- or post-impact fire.

The cabin, engine compartment, and main rotor mast were oriented approximately level in roll attitude. The forward cabin (from the front seats forward) was bent up relative to the aft cabin. The forward cabin was pitched up approximately 60 degrees, while the aft cabin remained approximately level in pitch. The cabin exhibited significant crush damage in the longitudinal direction. Both landing skids were fracture-separated from their forward struts.

The engine remained partially attached to its mounts, and exhibited crush damage to the baffles, accessories and other hardware installed on its aft and lower sides. The aft aspects of the engine and engine compartment exhibited crush damage in the forward direction. Clods of grass and dirt were embedded in the folds, openings and other deformed areas of those aft-facing components, and dirt was also packed into the spaces between the blades of the "squirrel cage" engine cooling fan. The lawn and ground immediately aft of and below the aft end of the helicopter engine compartment was disturbed with multiple scars and divots.

The main rotor mast remained affixed to the helicopter, and was oriented approximately upright. The two main rotor blades remained attached to the hub, and the outboard end of each blade was fracture-separated from the main segment of the blade. For reference purposes, the two blades were arbitrarily designated 'A' and 'B'. Blade A was found rotated in the hub so that it was nearly inverted, and its fracture-end was close to the ground. That blade was bent up (relative to normal blade orientation) approximately 90 degrees at a point about 32 inches from the blade root. The outboard 47 inches of the blade was fracture-separated parallel to the blade chord, but the separated segment was never found. Blade B was also bent about 90 degrees up, at a point about 22 inches from the root. The blade had cut into, and remained embedded in, the porch roof. The leading edge 'D' section was fractured through at a point about 49 inches from the tip. Aft of the 'D' section, the fracture extended inboard for about 17 inches, and then chordwise to the blade trailing edge. The 17-inch D section inboard of the fracture was curved down (relative to normal blade orientation) in an arc of about 90 degrees along its span. The liberated tip segment of blade B was found about 80 feet prior to the main wreckage. Spanwise, from the 90 degree bend point to the outboard separation fracture, each main blade was arced up (relative to normal blade orientation) to yield an approximate 20 degree change in the blade plane. The upper skin surfaces of both blades had multiple compression-type creases. Some creases were oriented chordwise, while others were oriented diagonally, typically offset about 45 or 60 degrees from the chordline.

The tail boom was partially fracture-separated just aft of the fuselage, and the forward 5-foot section of the tailboom was oriented approximately vertically, with the aft end pointing up. The aft segment of the tailboom, approximately 12 feet long, was folded down so that the tail rotor was in contact with the ground. Both tail rotor blades remained attached to the tail rotor gearbox, which remained attached to the tailboom. One blade was intact, and the other one was crumpled in the spanwise direction. The horizontal and upper vertical stabilizers were essentially intact, and the lower vertical stabilizer was crushed in the upward direction.

The helicopter sustained severe structural disruption in the region of the vee-belt sheaves, and the two sheaves were significantly displaced from their design orientation and location. All four vee-belts were found in the wreckage. Only one vee-belt remained in contact with the upper sheave, but all four were found in contact with the lower sheave. None of the vee-belts were found positioned in the grooves in the sheaves. No damage that could be associated with the failure or liberation of any vee-belt while operating at high speed was noted.

The carburetor heat control knob was unlocked and found extended about one-quarter of the total travel required to apply full carburetor heat. The carburetor heat control cable was intact from the cockpit to the airbox, and was free to move after the accident. The carburetor heat slider valve in the airbox was found crushed and bent in a position consistent with the application of full carburetor heat, and was not free to move after the accident.

The mixture control cable was separated from carburetor mixture control arm, but the mixture control arm spring remained in place. The carburetor bowl contained blue aviation gasoline which was clear and bright. The fuel was tested with water-detecting paste, and no water was detected in the sample. The main fuel tank remained attached to the fuselage, but was split open at a manufactured joint. The auxiliary tank was separated from the fuselage, and was breached. No fuel was observed in either tank, but both exhibited bulging consistent with hydraulic deformation. The finger screens for the main and auxiliary fuel tanks were clean and undamaged. The fuel strainer contained blue aviation gasoline which was clear and bright. The fuel was tested with water-detecting paste, and no water was detected in the sample. The strainer screen and bowl were free of contaminants.

A full set of flight controls was found to be installed at both the right and left seat positions. Control continuity for the cyclic, collective, and anti-torque systems was established; all separations were consistent with overload fractures. All separations in the tail rotor driveshaft were consistent with bending overload. Main and tail rotor gearbox continuity was confirmed. The magnetic chip detectors from the two gearboxes were free of debris. None of the "Telatemp" temperature decals indicated an over-temperature condition. The upper sheave overrunning clutch assembly functioned properly when rotated by hand.

With the exception of the 5-ampere "Clutch Start" circuit breaker, all circuit breakers in the panel located just forward of the left seat were in their normal (pushed in) positions.

Several components were removed from the wreckage for subsequent examination and testing. Hand rotation of the engine confirmed magneto impulse coupling function, and valve-train and crankshaft continuity. The engine was shipped to Lycoming for further examination. The maintenance facility in Mississippi reported that they had drained "seven quarts of clean oil" from the engine in preparation for shipment to the engine manufacturer.

According to first responders from the Jackson Fire Department, who arrived on scene about 7 minutes after the accident, both helicopter occupants initially survived the impact, but were in "critical condition." Both occupants were provided with "basic life support," extricated from the wreckage, and transported for additional medical care.

Pilot in Command

The PIC survived the accident, with serious injuries. According to FAA-provided information, the PIC suffered fractures in his legs, arm, and torso, but none in his feet, hands, or wrists.

Second Pilot

The second pilot died shortly after the accident, and the Mississippi State Medical Examiner office conducted the subsequent autopsy. According to the autopsy report, "evidence of emergency medical intervention includes an endotracheal tube, cervical collar, EKG pads, IV line, and hospital wrist band on the left wrist." The autopsy report noted fractures of the ribs, sternum and vertebrae, but none in the feet, hands or wrists. The autopsy report listed the cause of death as "multiple blunt force injuries."

The autopsy report stated that alcohol and screened drug test results were all negative, with the exception of caffeine detected in the blood. The Civil Aeromedical Institute (CAMI) toxicology report indicated that tests for carbon monoxide, cyanide, and ethanol were negative. Atropine was detected in the liver and blood, and naproxen was detected in the blood. A review of the second pilot's FAA medical records by the National Transportation Safety Board (NTSB) medical officer found that the documents contained "nothing of note."

N33PX Usage and Configuration Prior to the Accident

The owner of the helicopter stated that he was the last one to fly it prior to the accident flight. That flight was conducted on August 30, 2009, and lasted about 1 hour. According to the owner's records found in his hangar, after that flight, the owner topped off the helicopter with 25.3 gallons of fuel from his fuel storage tank at his hangar at MS90; he estimated that the total fuel on board at the beginning of the accident flight was 47 gallons. He did not report any problems with the helicopter. The owner also stated that he normally kept only the right-seat flight controls installed in the helicopter, and stowed the left-seat flight controls in a compartment under the left seat.

The owner stated that on the morning of September 1, the PIC telephoned him, and inquired about flying the helicopter that day. The owner told the PIC that the helicopter was available, and that it had full fuel, but that the PIC needed to check the engine oil quantity. The owner was not present at MS90 on the day of the accident. The owner stated that when he last saw the helicopter two days prior to the accident, all four doors were installed, but that subsequent to the accident, he found the helicopter's right front door in his hangar at MS90.

Radar-Derived Descent Rates

A total of three descents of the helicopter were captured on the ground-based radar. The radar data included altitude information, which enabled the calculation of descent rates. The helicopter's first approach to and landing at HKS was conducted with a descent rate of approximately 280 feet per minute (fpm). The second approach was conducted at approximately 450 fpm. The third (accident) descent occurred at a rate of approximately 2,400 fpm.

Fuel

Two days after the accident, the fuel in the owner's fuel storage and dispensing tank at MS90 was tested with water-detection paste; no water was detected in the fuel.

Cockpit Warning Lights

The helicopter was equipped with 13 warning lights in the cockpit that were designed to illuminate individually to alert the pilot to certain conditions. One (BRAKE) was mounted on the left side of the center control pedestal. Four more (ALT, ENG FIRE, OIL, GOV[ernor] OFF) were mounted in row on the upper right side of the central control pedestal, just below the flight instrument panel. Eight more lights (CLUTCH, MR [main rotor] TEMP, MR [main rotor] CHIP, CARBON MONOXIDE, STARTER ON, TR [tail rotor] CHIP, LOW FUEL, LOW RPM) were mounted in a row at the top center of the flight instrument panel.

The individual bulbs were removed and examined for indications of a phenomenon referred to as "hot filament stretch." According to a Transport Canada research paper, when illuminated, the filament of an incandescent light bulb operates at a temperature which renders it ductile. Hot filament stretch can occur when an operating incandescent light bulb is subjected to an inertial load from a significant acceleration such as an impact. The coiled filament distorts and/or stretches from the applied load, and the filament can take on an "uncoiled" appearance. The degree and type of filament distortion is a function of the filament orientation relative to the load direction, the support structure for the filament, and the magnitude, onset rate, and duration of the load. The bulb mounting structure, stiffness, and location will also affect the load parameters, and therefore the degree and type of filament distortion. A hot filament will not always distort during impact. A cold (unlit bulb) filament is brittle, will not distort when subjected to such loads, but may fracture. Only two conditions enable definitive conclusions about bulb status to be drawn from filament analysis; a distorted filament was heated at impact, and a fractured, undistorted filament was not heated at impact. Other conditions do not permit any definitive conclusions to be drawn.

With the exception of one bulb, all filaments were intact and undistorted. The filament from the "GOV OFF" bulb exhibited moderate localized deformation, but was not fractured.

Vee-Belt Inspection

After the accident, all four vee-belts were inspected in accordance with the Section 2.507 ("Vee-Belt Inspection") of the helicopter manufacturer's maintenance manual (MM) as guidance. The MM cited certain conditions that required replacement of the vee-belts, including cracking, fraying, blistering and cuts. Aside from the cuts that severed each vee-belt, two of the vee-belts exhibited damage sites that exceeded the "replace" threshold criteria in the MM. All damage sites were highly localized, and appeared to be consistent with impact damage.

Vee-Belt Tension Actuator Assembly

The vee-belt tension actuator assembly was sent to the helicopter manufacturer's facility for examination and testing. The recovered components consisted only of the upper half of the assembly. The components exhibited impact damage and deformation, and the motor sustained crush damage, and was encrusted with mud. Damage to the unit precluded performance of the manufacturer's standard acceptance test, so a functional check was performed instead. The functional check consisted of applying electrical power to the motor, and observing the extension and retraction of the actuator tube. The travel distance of the actuator tube was measured to be about 40 per cent of the normal travel, but it was observed to travel in either direction per the design. The current draw of the motor was measured to be within acceptable limits.

Lower Sheave Inspection

After the accident, the lower sheave was inspected in accordance with Section 2.508 ("Lower Sheave Vee-Belt Wear Pattern Inspection") of the MM; the sheave wear patterns were within the acceptable range specified by the MM. However, the aft-most land of the lower sheave exhibited impact damage along a 2-inch section on its aft face.

Upper Sheave, Clutch Assembly, and Clutch Shaft Inspections

The upper sheave, clutch assembly and clutch shaft were removed from the helicopter as a unit. The clutch assembly and upper sheave were inspected in accordance with the criteria in Section 2.410 ("Inspection Procedures and Checklist") of the MM. When viewed from the front (looking aft), the upper sheave was observed to be free to rotate counterclockwise on the shaft, but locked when rotated clockwise, causing the sheave and shaft to rotate together. With the exception of some impact marks, the clutch assembly and upper sheave were determined to be in compliance with the MM criteria. The Telatemp indicator on the upper actuator bearing was checked in accordance with Section 2.130 ("Telatemp Indicators") of the MM. The indicator showed that the highest indicated temperature was 180 degrees F, which was within acceptable limits.

Engine Examination and Test Runs

The engine was examined and run at the engine manufacturer's factory in Williamsport, Pennsylvania. The testing indicated that the engine operated normally, and was capable of developing its rated power.

Due to impact damage, four pushrod shroud tubes, one pushrod, and one rocker box cover were removed and replaced with undamaged components, and two ignition lead ends were repaired. The engine oil filter was removed, cut open and examined. Approximately six small black flakes were observed, and their appearance and mechanical characteristics were consistent with normal engine operation and service. The oil suction screen was mostly clean, but contained four small particles which were subsequently analyzed in the engine manufacturer's materials laboratory. The laboratory report indicated that one particle was consistent with engine combustion products, two were consistent with silicone rubber material, and the fourth was "a mineral material of unknown identity, possibly a dirt particle."

The timing of each of the two magnetos was within specified limits. The alternator was impact-damaged; it was removed because it was not needed for the engine test runs.

All spark plugs were removed and examined visually. All were consistent in appearance with normal wear. Although the electrodes of two spark plugs were oily, all were re-installed without cleaning or other remedial actions.

The installed oil filter adapter was removed and replaced with one with accommodations for the test cell instrumentation, a new oil filter was installed, and new oil was added. The accident starter would not turn the engine over; it was removed and replaced with another starter. The engine then started after approximately 20 to 30 seconds of cranking. The engine was operated under manual control for approximately 5 minutes, and then was transitioned to the computer-controlled production test program. The first full power attempt only reached an engine speed of 2,620 rpm, instead of the target value of 2,800 rpm. The low rpm was attributed to an incorrect test propeller pitch adjustment. Propeller pitch adjustments were made, and subsequent runs demonstrated that the engine achieved its "rated power" rpm. The engine was operated continuously for about 1 hour and 25 minutes, including 10 minutes at the "rated power" values.

The demonstrated rpm drop for each magneto was within limits. However, the difference (36.6 rpm) between the two rpm drops exceeded the limit value of 35 rpm. An engine manufacturer representative stated that the discrepancy could have been caused by the magneto timing, or the observed damage to one of the magneto condensers. The factory-specified lower limit for engine airflow was 1542.7 pounds per hour (pph), but the lowest measured airflow value was 1540.5 pph. An engine manufacturer representative advised that airflow could be modified by an additional adjustment of the propeller pitch.

Based on the observed engine run results, it was agreed that there was no reason to further examine and/or disassemble the engine. However, the following day, an additional test run in the same production test cell was conducted, in order to observe the engine response to rapid throttle movements. Unlike the R44, the test cell throttle linkage and control system was not a direct mechanical connection, and the test cell throttle rate was estimated to be approximately one-half that on the R44. After start and warm-up, approximately six each rapid throttle advances and retards were conducted. All parameters were normal throughout the throttle cycling, and the engine did not hesitate, stumble or exhibit any other abnormal characteristics during or after those cycles.

Right Magneto

Since the right magneto was used to provide the input signal for the engine tachometer, it was removed and shipped to its manufacturer, Teledyne Continental Motors (TCM), for testing. The magneto was tested in accordance with TCM production test criteria, and performed within acceptable limits.

Collective Control Assembly and Engine Governor Motor

The collective assembly was sent to the helicopter manufacturer's facility for examination and testing. The collective assembly consisted of the right-seat pilot's collective stick and throttle control, collective interconnect structure including the stub of the left-seat pilot's stick, and the governor motor and its clutch sub-assembly. The assembly was observed to be fracture-separated from the helicopter structure at the right-seat side, and was disassembled at its remaining structural mount points, throttle control outputs, and collective push-pull tube connecting rod.

A functional test was performed on the governor motor utilizing the manufacturer's test equipment. The motor rotated freely in both directions at all input values. The motor was driven to its full stop positions in both directions, and the clutch assembly was observed to function properly.

Engine Governor Control Unit

The electronic control unit for the engine governor was sent to the helicopter manufacturer's facility for examination and testing. Visual examination revealed a single dent in the top corner of the casing. The controller was connected to the facility test fixture, and tested in accordance with the production/quality test RPS-69 rev E. The controller passed all tests except sequence reference number 6.1.24, 6.1.38, and 6.1.39. These tests were related to the controller deadband zone that bracketed the target engine rpm; the deadband width exceeded the new production standards, but the controller was still able to maintain an engine within the target rpm values of 100.4 to 102.6 percent. Test 6.1.37 could not be performed due to its dependence on the results of 6.1.24, and was therefore categorized as a failure. The specific test results and parameters are attached to that test report.

After testing, the case lid was removed, and the internal components were visually examined. No indications of component failure, heat damage, or conductor contact with the casing were observed.

Latent Fingerprint Examination

At the request of the FAA, the circuit breaker panel and engine governor toggle switch were sent to the Federal Bureau of Investigation (FBI) laboratory in Quantico, Virginia for detection of latent fingerprints belonging to either of the two individuals on board the helicopter at the time of the accident. No latent fingerprints that could be positively identified as belonging to either of the individuals were detected on the toggle switch, the "Clutch Start" circuit breaker, or the area surrounding the circuit breaker.

POH Excerpts

The POH Normal Procedures section on autorotation practice stated that the governor "may be switched off if desired." The section also contained a CAUTION that stated "During simulated engine failures, a rapid decrease in rotor RPM will occur, requiring immediate lowering of collective control to avoid dangerously low rotor RPM. Catastrophic rotor stall could occur if rotor RPM ever drops below 80% plus 1% per 1000 feet of altitude."

Helicopter Manufacturer's Safety Notices

The helicopter manufacturer developed and published Safety Notices (SN) that were intended for R22 and R44 pilots. Each SN was numbered, and addressed a specific topic of flight or ground safety. All appropriate SNs were incorporated into the respective POH, and all were available on the manufacturer's website.

SN-20 ("Beware of Demonstration or Initial Training Flights") stated that "a disproportionate number" of accidents occurred during such flights "because individuals other than the pilot are allowed to manipulate the controls without being properly prepared or indoctrinated." The SN stated that an experienced flight instructor could easily regain control if the student did not make any "large or abrupt control movements," but warned that if the student made a "sudden large control movement in the wrong direction, even the most experienced instructor may not be able to recover control." It added that instructors "are seldom prepared for the student who loses control and does the wrong thing."

SN-24 ("Low RPM Rotor Stall Can Be Fatal") stated that "rotor stall due to low RPM causes a very high percentage of helicopter accidents." Rotor stall "can occur at any airspeed," which results in the cessation of lift production and "the aircraft literally falls out of the sky." Rotor stall results in a large drag increase, which causes the rotor RPM to "rapidly decrease, further increasing the rotor stall. As the helicopter begins to fall, the upward rushing air continues to increase the angle-of-attack…making recovery virtually impossible, even with full down collective."

SN-27 ("Surprise Throttle Chops Can Be Deadly") stated that "many flight instructors do not know how to give a student a simulated power failure safely," and therefore the instructor must be prepared to handle any unexpected student reaction. The SN advised instructors to "go through the exercise together a number of times until the student's reactions are both correct and predictable," and provided specific guidance for the execution of the exercises.

SN-29 ("Airplane Pilots High Risk When Flying Helicopters") cautioned that "the ingrained reactions of an experienced airplane pilot can be deadly when flying a helicopter." In particular, the SN warned that "application of forward stick when a pilot hears a horn (low RPM) would drive the RPM even lower and could result in rotor stall, especially if he also "adds power" (up collective). In less than one second the pilot could stall his rotor, causing the helicopter to fall out of the sky…The pilot does not have time to realize he made the wrong move, think about it, and then correct it. It's too late; the rotor has already stalled."

Carburetor Heat and Carburetor Icing

The R44 POH stated that when conditions conducive to carburetor ice were known or suspected to exist, pilots should apply carburetor heat to keep the carburetor air temperature (CAT) gauge needle out of the yellow arc. The Limitations section of the R44 POH indicated that the helicopter CAT gauge yellow arc extended from -19 degrees C to 3 degrees C. The POH contained a CAUTION that stated "the pilot may be unaware of carb ice formation as the governor will automatically increase throttle and maintain constant manifold pressure and RPM."

The POH stated that the accident helicopter was equipped with a "carb heat assist" mechanism that "correlates application of carburetor heat with changes in collective setting to reduce pilot workload." The system was designed to add carburetor heat when the collective was lowered, and vice-versa, and could be manually overridden. The carburetor heat control was also equipped with a latch that could be used to lock the carburetor heat control in the off position when not required. The POH recommended that the heat control was to be unlatched (and thereby enable the carburetor heat assist) whenever the temperature was less than 27 degrees C, and the spread between the temperature and dew point was less than 11 degrees C.

The helicopter manufacturer's SN-25 ("Carburetor Ice") stated that "carburetor ice can cause engine stoppage and is most likely to occur when there is high humidity or visible moisture and air temperature is below 70 degrees F (21 C)." SN-31 ("Governor Can Mask Carb Ice") stated that with the "throttle governor on, carb ice will not become apparent as a loss of either RPM or manifold pressure," and advised pilots that "when in doubt, apply carb heat as required to keep CAT out of yellow arc." It also stated that "if carb heat assist is used, it will reduce carb heat when you lift off to a hover and the control may require readjustment in flight."

An article by an R44 flight instructor who also conducted the R22/R44 pilot flight safety courses in Europe stated that carburetor icing in helicopters was "potentially more catastrophic" than icing in airplanes, since the airplane has the "benefit of airflow through the propeller that helps keep the engine turning," whereas the clutch system in the R44 prevents the rotor from driving the engine, which can lead to a "low rpm rotor stall."

When he was asked by the NTSB investigator-in-charge (IIC) about R44 carburetor icing potential, the local FAA Safety Team (FAAST) team manager stated that since Jackson is humid, "even if it's a clear day, it's a good idea to keep an eye on the carburetor air temperature gauge." He added that there is a strong potential for carburetor ice, and emphasized that "strict reliance on the specified procedures" and attention to the carburetor air inlet temperature gauge "is a necessity." He stated that he has had several instances of carburetor icing in the helicopter. He stated that the PIC was "acutely aware" of the potential for carburetor icing, that the PIC was knowledgeable about the phenomenon, its hazards, and preventive measures, and that the PIC "kept his eye" on the carburetor air inlet temperature gauge

The temperature and dew point values for the approximate time and location of the accident indicated that the relative humidity was approximately 45 percent. When the intersection of the two temperature values was located on a chart that depicted carburetor ice envelopes, the point was in the envelope labeled as "Serious Icing at Glide Power."

Special Federal Aviation Regulation 73

In 1995, the FAA issued SFAR 73, entitled "Robinson R22/R44 Special Training and Experience Requirements." According to SFAR 73, it applied "to all persons who seek to manipulate the controls" of a Robinson model R44 helicopter, and specified prerequisite training, aeronautical experience, endorsements, and flight review requirements. In part, SFAR 73 required that "no person may manipulate the controls of a Robinson mode... R44 helicopter for the purpose of flight unless" certain "awareness training" was completed, and the person's logbook was endorsed by a "certified flight instructor authorized under" SFAR 73.

The SFAR required that the awareness training "must be conducted by a certified flight instructor who has been endorsed under" SFAR 73, and the instruction topics included "energy management, mast bumping, low rotor RPM (blade stall), low G hazards, and rotor RPM decay." The SFAR also stated that a "person who can show satisfactory completion of the manufacturer's safety course after January 1, 1994, may obtain an endorsement from an FAA aviation safety inspector in lieu of completing the [specified] awareness training."

According to the helicopter manufacturer's records, the second pilot had not attended or completed the manufacturer's safety course.

When he was asked by the IIC about the second pilot's status regarding compliance with SFAR 73, the Jackson FSDO manager stated that the second pilot had not received the required training or endorsements, and therefore was not in compliance with the requirements of SFAR 73. He stated that SFAR 73 did not yet apply to the second pilot, since the second pilot would not begin training in any Robinson helicopters until he was in compliance with SFAR 73. According to the FSDO manager, the second pilot was "not authorized" to conduct "4040 flights" in the Robinson; there was no FAA paperwork approving him to do so. The manager stated that each FAA employee participating in the 4040 program had to be signed off by a designated instructor in order to operate a particular aircraft.

When he was asked by the IIC about the second pilot's status regarding compliance with SFAR 73, the FAAST team manager stated that several days prior to the accident flight, he was aware that the second pilot had no R44 time, and probably did not meet the requirements of SFAR 73. The manager stated that as of the day of the accident flight, he did not know whether the second pilot had satisfied SFAR 73 requirements for the flight, but added that it was not his responsibility to ascertain or ensure that compliance.

The SFAR also stated that no certificated flight instructor could provide instruction or conduct a flight review in an R44 unless that instructor completed the specified awareness training; had at least 200 flight hours in helicopters, including at least 50 hours in the R44; completed flight training in a Robinson helicopter on abnormal and emergency procedures that included enhanced training in autorotation procedures, engine rotor RPM control without the use of the governor, low rotor RPM recognition and recovery, effects of low G maneuvers and their proper recovery procedures; and had been authorized by endorsement from an FAA aviation safety inspector or authorized designated examiner that the instructor had completed the appropriate training, met the experience requirements, and had satisfactorily demonstrated an ability to provide instruction on the general subject areas and flight training specified by SFAR 73.

FAA Flight Program

The Flight Standards Service of the FAA had a formal nationwide "flight program" intended to ensure that FAA inspectors met the "skill, tasks, and formal training requirements that are deemed essential to the currency and proficiency" of the flight program participants. The program guidance was contained in FAA Order 4040, and the program was informally referred to as the "4040 program." The stated rationale for the program was that the "interest of the public, the safety of the workforce, and the credibility of the organization are best served by having qualified, proficient, and current inspectors conducting pilot evaluating, testing, and checking functions."

According to the FAA, the Jackson FSDO was allocated $41,398.21 for the program's 2009 budget, and Jackson FSDO personnel had flown 73.3 hours. No breakdowns of individual personnel hours or aircraft types were provided. FAA records indicated that Jackson FSDO personnel who participated in the program did not have any accidents or incidents other than the subject accident.

Chapter 3 Section 19 of the "Mississippi FSDO Office Policy" document provided additional guidance on the execution of the flight program. The guidance stated that prior to the first week of each quarter, the participating inspectors and their supervisors had the "joint responsibility to prepare and submit a flight schedule" on FAA Form 4040-6 (FAA Aircraft Request and Use Record). The guidance stated that the "schedule will utilize the first two months of each quarter to accomplish each quarter's flight time," and that "the supervisor must approve any flight schedule changes by initialing the 4040-6" near the revised flight date. It also stated that "All flights shall have a properly completed and approved FAA Form 4040-6," that "the manager will provide approval for each flight by signing each FAA Form 4040-6 just prior to the planned flight," and that a "flight plan will be filed for each flight." The Form 4040-6 for the PIC that was provided to the NTSB did not conform to any of these requirements, and no flight plan was filed for the accident flight. No Form 4040-6 for the second pilot was provided to the NTSB.

The Jackson FSDO manager was responsible for the administration of the facility's 4040 program. In an interview with the IIC, he stated that three other persons at the FSDO assisted him with his program duties, including the FAAST team manager who functioned as the program safety officer. The FSDO manager estimated that he spent approximately 1 week per quarter, including his attendance at the quarterly safety meetings, administering the 4040 program. As of date of accident, three personnel in the FSDO were participating (actively flying) in the 4040 program; two of them were the pilots involved in the accident. The FAAST team manager also flew actively in the program, but he was not included in the FSDO manager's count.

The FSDO manager stated that the local procedures required participating pilots to fill out a form (Form 4040-6) and submit it to the FSDO manager, who authorized pilots to conduct the flights by virtue of his signature on the form. The FSDO manager stated that pilots filled the forms out on a quarterly basis, and that the most recent quarter prior to the accident encompassed June to August. He stated that the funding that was allocated for that quarter was not utilized by the end of the quarter, and the FSDO requested and received formal approval from the FAA Regional Office ("ASW-240") in Dallas, Texas to use the remaining money to permit pilots to fly in September 2009, but that the 4040-6 was not updated to reflect the extension.

A computer program known as the "Flight Activity and Crew Tracking System" (FACTS) was used to track pilots' currency, qualifications and compliance with all applicable policies, practices, and Federal Aviation Regulations. The FACTS program was designed to alert the FSDO manager and others of pending and current discrepancies, in order to ensure that appropriate preventative or corrective actions could be taken by the responsible personnel.

According to the FSDO manager, the Safety Officer was the individual who typically briefed and/or indoctrinated FAA new hires regarding the 4040 program.

A few weeks after the accident, when he was asked by the IIC what changes, either temporary or permanent, were made to 4040 program guidance, execution, or personnel subsequent to the accident, the FAAST team manager stated that there was limited 4040 flight activity after the accident, but that he was not aware of any changes or pending changes as a result of, or subsequent to, the accident.

In March 2011, the FAA formally responded to an NTSB inquiry regarding post-accident changes to the 4040 program. The response letter stated that in late 2010, the FAA had formed an internal Helicopter Safety Work Group that was tasked with evaluating the helicopter flight program within the FAA. As a result of that Task Force, the FAA made numerous changes to the program, which were enumerated and summarized in the response letter.

Knowledge of Jackson FSDO Personnel Regarding Accident Flight

In an interview with the IIC, the Jackson FSDO manager was asked what he knew, prior to the flight, about the accident pilots and the accident flight. He stated that the second pilot was hired primarily for his helicopter experience. He also stated that the second pilot was not qualified in Robinson helicopters, but "the goal [was] to get him [qualified] in the future." The FSDO manager stated that he informed the PIC that the second pilot was not to manipulate the controls of the Robinson on this flight, since the second pilot was not in compliance with SFAR 73. When asked whether he had any particular concerns regarding the flight, due to the facts that the PIC was a long-term FAA employee and a highly-experienced pilot, but with low time in helicopters, and that the second pilot was new to the FAA, but a highly experienced helicopter pilot, the FSDO manager stated that he had no concerns; it was to be a "familiarization" flight, in which the second pilot would not manipulate the controls. The FSDO manager stated that he did speak to the second pilot about the upcoming flight, but only briefly, and the FSDO manager did not provide any specifics of the conversation. He also stated that to his knowledge, no-one else in the FSDO had any detailed knowledge about the preparations for the flight. No other Jackson FSDO personnel with knowledge of the flight were identified by the investigation.

When the FAAST team manager was asked the same series of questions by the IIC, he stated that he was aware of the experience discrepancy, but that it was not a concern to him. He did not speak to the second pilot the morning of the accident flight. The FAAST team manager stated that the PIC asked him which pilot should take which seat in the helicopter, and that the FAAST team manager recommended that the PIC fly from the right seat, due to his recency of experience, and his higher time and greater familiarity with flying that helicopter from the right seat. The FAAST team manager stated that the PIC quickly concurred, and that he had the sense that the PIC was simply seeking affirmation of that seating arrangement. They did not discuss the status of the installation of flight controls for the accident flight. The manager stated that he was familiar with the PIC's flying habits and capabilities, and therefore he had no particular concerns about the conduct or potential hazards of the flight. The FAAST team manager stated that aside from the seat discussion, he did not discuss the flight with either of the pilots, and he did not discuss the flight with the FSDO manager.

The FAAST team manager was not aware that the FSDO manager told the PIC that the second pilot was not to manipulate the controls of the helicopter. The FAAST team manager added that the term "familiarization flight" was not formally defined by the FAA, and that his personal version of a familiarization flight would consist of exposing the individual to the aircraft, as well as the airspace, airports and personnel in the local area. The FAAST team manager was not aware of any other individuals at the FSDO who had knowledge of the flight. He stated that the normal procedure for a pilot who was planning to conduct a flight was that the pilot would inform another individual in the FSDO (typically another Operations inspector) about the planned flight, and include basic information such as departure time, flight route or destination(s), expected maneuvers and exercises, and a return time. The FAAST team manager stated that the information transfer was typically accomplished in one of two ways; either via a paper copy of a "flight planning sheet," or electronically via the "X-out" program. No records of either of those methods regarding the accident flight were provided to the NTSB.