NTSB Identification: ERA11FA272 14 CFR Part 91: General Aviation Accident occurred Saturday, April 30, 2011 in Indiana, PA Probable Cause Approval Date: 08/13/2014 Aircraft: ROBINSON HELICOPTER COMPANY R44 II, registration: N445AB Injuries: 1 Fatal, 3 Serious. NTSB investigators either traveled in support of this investigation or conducted a significant amount of investigative work without any travel, and used data obtained from various sources to prepare this aircraft accident report. The single-engine helicopter was operating in an urban area with a three-person film crew onboard who were following the movements of police vehicles. One of the passengers reported that, while following cars and filming them, the pilot turned the helicopter about 180 degrees while not moving in a forward direction. The helicopter then started bucking forward and backward and, at the same time, began to spin downward until it impacted a building. A witness reported that he saw the helicopter flying about 100 feet above ground level at a very slow rate of forward airspeed in a west-to-east pattern while making multiple passes. He said that the helicopter was making 180-degree flat turns (without a banking attitude). A few minutes later, he saw the helicopter flying in a northwesterly direction, enter what he believed to be a short sudden autorotation, and then lose altitude. Postaccident examination of the helicopter did not reveal any preexisting mechanical anomalies with the airframe, drive system, or engine that could have contributed to the accident. Two segments of video were recovered from one of the video cameras onboard the helicopter. The second segment was 73 seconds long and contained the final moments of the accident flight. Sixty-one seconds into the segment, the camera zoomed out, and the helicopter began a right rotation, before which, there was very little evidence of forward airspeed. Between 61 and 73 seconds, the helicopter completed two full right rotations. At 63 seconds, the helicopter pitched nose down, and, at 67 seconds, it then returned to nose level and lost altitude. At 68 seconds, the helicopter pitched nose up for 3 seconds. At 71 seconds, it pitched further nose up and then completed a quarter of a rotation before the video ended. From 67 seconds to the end of the recording, the main rotor blades came into view, and rotor blur was visible. Given the witness and video evidence, the helicopter likely experienced a loss of translational lift and corresponding loss of sufficient tail rotor authority to maintain level flight because the pilot was operating the helicopter out of ground effect at a low forward airspeed. In March 1999, Robinson Helicopter issued Safety Notice SN-34 to be included as a supplement in the R-44 Pilot Operating Handbook; a revision was issued in April 2009. The Safety Notice indicated, in part, that, Aerial survey and photo flights should only be conducted by well trained, experienced pilots who Have at least 500 hours pilotin-command in helicopters and over 100 hours in the model flown. The accident pilot had about 388 hours of total helicopter flight time, including about 41 hours in the accident model helicopter. The National Transportation Safety Board determines the probable cause(s) of this accident to be: The pilot's operation of the helicopter at a low-forward airspeed out of ground effect, which resulted in the helicopter s loss of translational lift and tail rotor authority and the pilot s subsequent loss of helicopter control. Full narrative available HISTORY OF FLIGHT ERA11FA272 On April 30, 2011, about 2030 eastern daylight time, a Robinson R-44 II helicopter, N445AB, registered to Penn Helicopters LLC, Freidens, Pennsylvania, was substantially damaged following a collision with buildings in an urban section of Indiana, Pennsylvania. The commercial pilot sustained serious injuries, 1 passenger sustained serious injuries, and 1 passenger sustained fatal injuries. The helicopter was being operated under local contract as an aerial filming flight under the provisions of 14 Code of Federal Regulations Part 91. Visual meteorological conditions prevailed and a flight plan was not filed. The flight originated from the Indiana County Airport (Jimmy Stewart Field), Indiana, Pennsylvania, about 1930.
The helicopter was being flown with a three person filming contingent onboard. The purpose of the filming was to capture real time police activities. In particular, the helicopter was following the movements of police vehicles on the streets of Indiana, near the campus of Indiana University of Pennsylvania. Several witnesses reported seeing the helicopter operating at a low altitude above the city streets. They characterized the helicopter as flying slowly and making 90-degree and 180-degree turns while in a level attitude. One of the passengers reported the following: After takeoff, the pilot flew the helicopter around the city for two laps at an altitude of approximately 1,000 feet. After the laps around the city, the helicopter was flown to a lower altitude to facilitate filming of police cars. While following over top filming the cars, the helicopter turned about 180- degrees while not moving in a forward direction. The helicopter then started bucking (forward and backward motion), and at the same time, began to spin downward. A person who was out to dinner in the vicinity of the accident area reported that he saw the helicopter flying about 100 feet AGL at a very slow rate of forward airspeed in a west to east pattern making multiple passes. He said that the helicopter was making 180-degree flat turns (without a banking attitude) and then flying toward the center of town. He said that he told his dinner companion that he thought the helicopter was well below a safe altitude for any emergency procedure and that he was concerned for the safety of the crew of the aircraft. He said that he tried to call the Indiana County Airport at 20:31 to notify them of the low flying helicopter. A few minutes later, he saw the helicopter flying in a north westerly direction and then saw what he believed to be a short sudden autorotation whereby the helicopter lost altitude. He said that the nose of the helicopter pitched up and the tail went down. He estimated the forward airspeed was under 20 knots and the helicopter was about 100 feet AGL. He thought that the helicopter had recovered as it went out of his view and he did not hear the noise of the engine anymore. About 2-3 minutes later he was informed that there was a helicopter crash. He proceeded to the accident site to offer assistance to first responders. In NTSB Form 6120, submitted by the pilot, he stated that he had no memory of the accident sequence due to head injuries. There were no reported radio or distress calls from the helicopter before the accident. The helicopter was operating below the threshold for the nearest radar stations, therefore no radar data was available for the accident flight. First responders from the Indiana city police and fire departments took charge of the accident site and expeditiously transported the 4 occupants of the helicopter to a local hospital. There were no reported injuries to persons on the ground. PERSONNEL INFORMATION The pilot held commercial certificates for Multi-Engine Airplane (MEL), single-engine airplane (SEL), Rotorcraft Helicopter, and Instrument Airplane. He also held flight instructor certificates for Single-Engine Airplane and Rotorcraft Helicopter. He held a valid second class FAA medical certificate, dated June 17, 2010. The pilot's total flight hours were 1,492 (all aircraft) as reported on the submitted NTSB Form 6120, of which, 388 hours were in rotorcraft. A review of the pilot's logbooks showed about 347 hours in the Robinson R-22 model helicopter and about 40.5 hours in the Robinson R-44 model helicopter. AIRCRAFT INFORMATION 2005 model Robinson R-44 II had about 276.9 hours at the time of the accident. The helicopter was configured with a single set of flight controls at the pilot station. The installed Lycoming IO 540 engine (original equipment) had about 279.0 hours. Calculated weight and balance showed that the helicopter was within flight operating limits for the accident flight. METEOROLOGICAL INFORMATION
Local witnesses stated that the weather was mild with little to no wind. The nearest weather reporting station was located at the Indiana County Airport IDI, about 10 miles from the accident site. The reported weather METAR at 2015 local time was clear, wing from 160 degrees at 4 knots, and 10 miles visibility. METAR KIDI 010015Z AUTO 00000KT 10SM 16/04 3014 RMK A02 FLIGHT RECORDERS The helicopter was not equipped with hardened recording devices. However, a professional video camera was recovered from the wreckage and sent to the NTSB Vehicle Recorder Laboratory, Washington DC, for evaluation. WRECKAGE AND IMPACT INFORMATION General on Site The accident site was confined to an urban area amidst several 2-story buildings. The helicopter struck the tops of 2 buildings and came to rest along the side of one building in a nose down position. The accident site was confined to an urban area amidst several two story buildings. The helicopter struck the tops of the two buildings and came to rest along the side of one building in a nose down position, with the cabin section crushed upward on the left side. The first impact points were identified on the top edges of the two buildings, and were consistent with main rotor blade strikes. One main rotor blade was found sheared near its blade root and was located on the roof of the white building. The other main rotor blade showed impact damage along its leading edge corresponding to scars located on the red brick building. The tail rotor blades showed damage consistent with side load impacts. The cabin section was crushed upward on the left side. The first impact points were identified on the top edges of the two buildings. The impact points were consistent with main rotor blade strikes. The helicopter was transported to a hangar at the Indiana County Airport for detailed examination. Airframe Structure and Cabin The airframe appeared to have impacted in an extreme nose low, left skid low attitude. The entire fuselage and all of the cowlings were deformed. The mast fairing sustained impact damage to its leading edge and was deformed around the lower edge. There was no evidence of contact between the main rotor blades and the tailboom. The upper vertical stabilizer had white, yellow and black scuff marks on its left side at the leading edge, running forward and aft. It was bent to the right approximately 35. The tail rotor guard and tail skid were not damaged. Parts of the roof, door posts, windshield bow, cyclic stick, seat back supports, seat bottom support structures and some of the seat belts had been cut by first responders to extract the occupants. The upper cockpit console was separated from the lower console and tethered by wiring and tubing. The lower console was deformed and displaced toward the left. All of the removable controls were located (stowed) under the forward left seat. The collective control was found in the full up position. The collective friction slider was bent at the lower mount about mid travel. The anti-torque pedals were found in a neutral position. The forward right seat belt was found unbuckled and the shoulder harness was cut. The forward left seat belt was buckled and the belt was cut. The aft left seat belt was buckled and the belt was cut. The aft right seat belt was unbuckled. All four seat support structures were deformed downward and forward. The forward left door was not installed at the time of the accident. The aft left door had moderate impact damage, the right side doors had minor impact damage, and the chin bubble and left side of fuselage showed significant impact deformations. All 4 seat supports were crushed and deformed.
The upper vertical stabilizer was bent and had scrape marks and paint transfer marks matching the color of the building. Landing Gear The left skid and struts were found pushed aft rotating the cross tubes in the elbows. The toe section was found disconnected from the left skid tube and the forward strut. The aft strut was found disconnected from the lower frame and partially separated from the skid tube. Both cross tubes appeared to be straight. No abnormal scoring or paint transfer marks were observed on either skid tube. Flight Controls The flight control system had discontinuities due to impact forces. There was no evidence of pre-impact failures or disconnects of the flight control system. The tail rotor pitch change slider was free to slide along the tail rotor gearbox output shaft. Fuel System At the time of the examination, there was no fuel remaining in the primary and auxiliary fuel tanks. The FAA reported that fuel samples were taken at the scene before the helicopter was transported and appeared to be clean 100LL aviation fuel. The main fuel tank was found separated from the cabin bulkhead and had impact damage to the inboard skin and the upper corner of the outer skin. The fuel cap was secure. The vent hoses were disconnected and were clear of blockages. The fuel feed and crossover lines were cut and/or disconnected for removal. The line B-nuts were found secured. The finger strainer was removed and found clear of blockages. The auxiliary fuel tank had minor impact damage. The fuel cap was secure. The vent hoses were disconnected and were clear of blockages. The fuel line B-nuts were found secured. The finger strainer was removed and found clear of blockages. The main fuel valve was found in the 'ON' position. There was no indication of fuel leakage prior to impact. The auxiliary fuel pump operated properly. The gascolator bowl was found full of what appeared and smelled like 100LL aviation fuel. No water was present. The gascolator screen was removed and found clear of debris. Engine Controls The fuel mixture control knob was found in the full rich position and the fuel mixture knob guard was found not attached to its retaining chain. The fuel mixture control arm on the fuel control servo was found in the full rich position and the fuel mixture control wire housing was secure on the support bracket. The fuel mixture control operated smoothly by hand. Main Rotor / Tail Rotor Systems The main rotor blade droop and teeter stops were found undamaged. There were rotational score marks on each end of the hub adjacent to the blade pitch boots. Main rotor blade, serial number 1698B, was bent upward near the root. Damage along the span of the blade was consistent with impact damage. Both the upper and lower surfaces had scuff marks running chord-wise and the leading edge had areas of white and blue paint transfer marks. Main rotor blade, serial number 1694B, was bent upward near the root. Damage along the span was consistent with impact damage. The spar was bent downward about 90 degrees about four feet inboard of the tip and bent several more degrees closer to the tip. The upper and lower skins were peeled away from the spar and there was multi directional scuff marks and white and blue paint transfer marks.
The flex plate of the intermediate flex coupling was found bent. The tail rotor driveshaft damper bearing rotated smoothly. The aft flex coupling was found not damaged. The tail rotor gearbox output shaft rotated freely with no anomalies and oil was visible in the sight gauge. Note: [The tail rotor blades were marked "A" and "B" for reference during the examination] Tail rotor blade "A" was bent inward near the root and had white and blue paint transfer marks on the outer surface at the leading edge and on the outer surface of the hub. Tail rotor blade "B" was bent in an S shape and the hub was bent inward. Blade "B" had white and blue and red paint transfer marks on the outer surface mostly near the leading edge. Blade "B" had a puncture in the outer skin near the tailing edge about mid span. Drive Line Continuity was established from upper drive sheave to main and tail rotors. Note: [The V-belts were numbered #1 through #4, by the investigative team for reference only] V-belt #1 remained on the upper sheave and was positioned forward of the lower sheave and was tensioned. This V- belt had a ridge on the backing, running lengthwise, and had localized abrasions on the edges in the area where the V-belt was forced between the forward edge of the lower sheave and the aft face of the flywheel. The aft face of the flywheel and the forward face of the sheave had rubber residue adjacent to where the V-belt was located. The #1 V- belt also retained curvature consistent with the arc of the flywheel. V-belts #2, #3, and #4 were aft of the upper sheave and not tensioned. The #2 belt had damage to the edge approximately 3 inches long. Belts #3 and #4 were not damaged. The observed displacement of the V-belts appeared to be the result of downward impact forces. The damage to the V-belts and the rubber residue on the sheave and flywheel appear to be the result of the stationary location of the V-belt while the engine was still running during impact and for a short time after impact. The upper sheave had rotational scoring on the forward face. The forward and intermediate flex couplings were bent. No pre-impact anomalies of the V-belt drive system were found. Engine Examination After transport to the hangar from the accident site, the engine remained attached to the helicopter by the engine mount. The engine cowling was impact damaged. The engine exhaust tailpipe was impact damaged. The fuel servo throttle plate was partially open and the mixture control lever was observed at about 2/3 rich. Rotational scoring marks on the engine cooling scroll, on the scroll shroud and on tubular structure forward of the scroll were consistent with engine rotation during the impact sequence. Since the helicopter came to rest partially inverted at the accident site, the lower spark plugs were removed and oil was allowed to drain from the right side cylinders. The aircraft fuel gascolator bowl was removed and contained a blue liquid with and odor consistent with that of aviation gasoline. No debris was noted on the gascolator fuel screen. The spark plugs were reinstalled and the helicopter was moved outside. Blocks were placed under the left side tubular mount structure to allow the aircraft to sit in a more level attitude. The fuel servo throttle and mixture levers could not be moved via the cockpit controls because of impact damage, so the control cables were cut near the fuel servo to allow the levers to be moved manually. The magneto switch was placed in the "both" position, the mixture lever was placed in the full rich position and the throttle control was placed in the idle position. A fuel line was connected to the inlet to the aircraft electric fuel pump and placed in a container of aviation gasoline. Jumper cables were connected to the starter, the electric fuel pump and an external 24 volt battery. The engine was primed using the aircraft electric fuel pump, started and allowed to run for about 3-4 minutes. The magneto switch was actuated and the engine observed to run in the "left",
"right" and "both" positions. The engine was stopped using the mixture control lever and the aircraft was moved back into the hangar for examination. The engine driven fuel pump, fuel injector servo, fuel flow divider and the fuel injector nozzles remained attached to the engine and no damage was noted. The fuel servo fuel inlet screen was removed and a small amount of dark colored material observed on the inside end of the screen. Both magnetos remained attached to the engine and no damage was noted. Engine operation was confirmed with the magneto switch in the "left, "right" and "both" positions. The lower spark plugs were removed. The electrodes exhibited gray coloration and worn normal condition. The #2, #4 and #6 lower spark plugs electrodes were oily. The upper spark plugs were not removed. No damage was to the ignition harness. The starter and alternator remained attached to the engine and no damage was noted. The engine was not equipped with a vacuum pump. The oil cooler remained attached and no damage or leakage was observed. The engine oil filter was removed and examined. No debris was noted in the folds of the filter media. No indication of pre-impact damage to or malfunction of the engine was observed. TESTS AND RESEARCH An On-Board Image Recorder Group was convened consisting of an NTSB recorder specialist, and FAA flight data specialist, and a representative from Robinson Helicopters. There were two segments of video that were recovered. The first segment was about 35 minutes long and the second segment was about 73 seconds long. The final moments of the accident flight were recorded on the 73-second segment. At 61 seconds into the second segment, the camera zoomed out and the helicopter began a rotation to the right. Just before the rotation began, there was very little evidence of forward airspeed. Between 61 and 73 seconds, the helicopter had completed 2 full rotations to the right. At 63 seconds, the helicopter pitched nose down, returning to nose level at 67 seconds. The first indication of altitude loss occurred at 67 seconds. At 68 seconds, the helicopter pitched nose up for three seconds before pitching further nose up at 71 seconds, completing a quarter of a rotation before the video ended at 73 seconds into the second segment. At 67 seconds, until the end of the recording, the main rotor blades came into view and rotor blur was visible. The entire On-Board Recorder Group Report is available in the public docket of this report. ADDITIONAL INFORMATION Robinson had issued a Safety Notice SN-34, included as a supplement in the R-44 Pilot Operating Handbook. It was originally issued in March of 1999 and revised in April of 2009. The entire SN-34 reads as follows: "AERIAL SURVEY AND PHOTO FLIGHTS - VERY HIGH RISK There is a misconception that aerial survey and photo flights can be flown safely by low time pilots. Not true. There have been numerous fatal accidents during aerial survey and photo flights, including several involving Robinson helicopters. Often, to please the observer or photographer, an inexperienced pilot will slow the helicopter to less than 30 KIAS and then attempt to maneuver for the best viewing angle. While maneuvering, the pilot may lose track of airspeed and wind conditions. The helicopter can rapidly lose translational lift and begin to settle. An inexperienced pilot may raise the collective to stop the descent. This can reduce RPM thereby reducing power available and causing an even greater descent rate and further loss of RPM. Rolling on throttle will increase rotor torque but not power available due to the low RPM. Because tail rotor thrust is proportional to the square of RPM, if the RPM drops below 80% nearly one-half of the tail rotor thrust is lost and the helicopter will rotate nose right. Suddenly the decreasing RPM also causes the main rotor to stall and the helicopter falls rapidly while continuing to rotate. The resulting impact is usually fatal. Aerial survey and photo flights should only be conducted by well trained, experienced pilots who:
1) Have at least 500 hours pilot-in-command in helicopters and over 100 hours in the model flown; 2) Have extensive training in both low RPM and settling-with-power recovery techniques; 3) Are willing to say no to the observer or photographer and only fly the aircraft at speeds, altitudes, and wind angles that are safe and allow good escape routes."