NTSB Factual Report

HISTORY OF FLIGHT

On December 20, 2003, at 1530 Pacific standard time, a Robinson R22 Beta, N7189U, lost engine power while in a climb, and made a forced landing in uneven mountainous terrain and rolled over in a creek bed about 12 miles east of the French Valley Airport (F70), Murrieta/Temecula, California. Sunstone Aviation, LLC, d.b.a. USA Academy of Aviation operated the rental helicopter under the provisions of 14 CFR Part 91. The helicopter sustained substantial damage. The pilot received minor injuries, and the passenger was seriously injured. Visual meteorological conditions prevailed for the local area flight that departed F70 about 1445, and a flight plan had not been filed. The flight was destined for F70.

According to the pilot, prior to leaving the Los Angeles area he checked the weather and received a standard weather briefing for the time around 1430-1700. He arrived at the airport at 1425, and picked up the helicopter keys from the Academy. While at the Academy, his flight instructor told him about an airplane doing aerobatics near Lake Skinner, and gave him information on weather and fuel. The pilot did a preflight inspection of the helicopter. He and his passenger got in the helicopter and he went through the start up checklist. Between 1440 and 1450, he noted that all of the engine gauges were in the green, the warning and cautionary lights were off, and the carburetor heat button was "pulled up" or in the ON position. He checked the windsock and observed that the winds were from 270 degrees at 5 to 6 knots. He then hover taxied to runway 18 for departure, with no abnormal readings noted from the instruments.

After departing F70, he flew to the south towards Lake Skinner where he did a couple of turns. Then he headed towards Vail Lake. The pilot reported that they flew in the Vail Lake area for about 30 minutes, where he conducted an in ground effect hover and out of ground effect hover (550 feet above ground level (agl)) around an island in the middle of Vail Lake. The pilot stated that he did this as a normal part of his routine as an extra precaution before he entered the canyon where the accident occurred. The pilot stated that all of the indications were normal and the helicopter was performing as usual.

As they flew over the mountains toward the canyon (accident location), he pointed out the scenery to his passenger. Once they reached the canyon he slowed the helicopter to about 35 - 40 knots indicated airspeed (KIAS). He noted that all of the gauges were in the green, the warning and cautionary lights were off, and the carburetor heat "was in good standing." He stated that he always checked the carburetor heat in the accident helicopter because the button "creeps down fast;" he further reported that he informed the Academy of this condition. His flight instructor told him that it was okay to fly the helicopter in that condition. The pilot stated that everything appeared normal at that point.

The pilot stated that he entered the canyon for the fifth time. The winds were between 0 to 5 KIAS, with no turbulence. He turned to the right towards a fork in the canyon that he wanted to show his passenger. He rechecked all of the gauges, with no abnormal indications. The helicopter performed normally, and he maintained about 150 feet agl in the canyon while performing a slow climb. The pilot stated that he had just finished checking the gauges, with no abnormalities, when the engine started to sputter. The pilot reported that the helicopter went from a climb to a "drop in power with power fluctuation." He immediately entered an autorotation, and checked for carb ice and checked "with my left hand where was the carb[uretor] heat button." He stated that there was no indication that the engine failure was due to carburetor ice at the gauge.

The pilot continued the autorotation to the ground, while trying to power up the engine. He stated that the engine did not have enough power, "nor did it sound like usual." The pilot stated that he maintained engine rpm (revolutions per minute), along with rotor rpm. He told his passenger that they "were going down."

The pilot reported that during the autorotation he did not see a landing area and committed the landing towards the right side of the canyon wall, which had an embankment about 5 to 10 feet above the canyon floor. He flared, leveled out the helicopter, and brought it to a complete stop. The helicopter touched down right skid first with the left skid about 5 to 10 feet above the ground. The pilot reported that he was worried about dynamic rollover from landing in an uneven attitude, but felt there was not enough power for the helicopter to hover. He stated that the collective was "pulled up but there was no real power from what I could tell, the engine/rotor system was still turning under partial power and not damaged."

The pilot stated that the helicopter slipped down the mountainside. After exiting the helicopter, he turned off the master switch due to leaking fuel. He and his passenger walked about 1/4-mile downstream from the helicopter for safety reasons. He then made a series of mayday calls until the sheriff's department air unit rescued them.

According to the passenger, the pilot called for weather and told her it was okay to fly. Once they arrived at F70, the pilot spoke with his flight instructor and then checked the helicopter. The passenger reported that the headset she was wearing was intermittent throughout the entire flight. She reported that it was not a sunny day, but the weather seemed fair. They flew by a lake, and the pilot "lowered the helicopter without landing - and then lifted it up again to the same height as before." After that they flew around for about 15 minutes, and everything seemed okay. At some point they entered the canyon to look at the view, and the pilot indicated that he had been there several times before.

The passenger stated that they were in the canyon a few minutes, and she was looking outside of her window (left side). When she turned to look in front of her she saw a mountain. She stated that it felt as if the mountain was coming towards them, and she realized something was wrong. She stated that instead of going up, they were going down. She turned to look at the pilot and the helicopter collided with a mountain. After hitting the mountain from the right side, the helicopter fell onto its left side and she recalls being stuck underneath the helicopter.

The passenger further reported that prior to the accident, the pilot told her that he had heard a noise and that "he felt the power was dying," and told her they were going down. She stated that she did not hear the noise that the pilot was referring to.

A Federal Aviation Administration (FAA) inspector traveled to the accident site, and reported that the canyon walls that surrounded the accident site were about 500 feet high. He observed the helicopter lying on its left side in a small creek bed with water running through it.

AIRCRAFT INFORMATION

According to the helicopter manufacturer's Pilot's Operating Handbook (POH) in section 4 - Normal Procedures under the section titled "Use of Carburetor Heat" in the CAUTION it states:

"The pilot may be unaware of carburetor ice formation as the governor will automatically increase throttle and maintain constant manifold pressure and RPM. Therefore, the pilot must apply carburetor heat as required whenever icing conditions are suspected."

At power settings above 18 inches manifold pressure (MAP), the pilot should apply carburetor heat as required to keep the CAT gauge needle out of the yellow arc. However, because the CAT gauge does not give a correct indication of carburetor temperature with power settings below 18 inches MAP, the pilot should ignore the gauge and apply full carburetor heat.

Also under section 4 - Normal Procedures in the section titled "Use of Carb[uretor] Heat Assist" it states:

A carburetor heat assist device is installed on R22s with O-360 engines. The carburetor heat assist correlates application of carburetor heat with changes in collective setting to reduce pilot workload. Lowering collective mechanically adds heat and raising collective reduces heat. Collective input is transmitted through a friction clutch, which allows the pilot to override the system and increase or decrease heat as required. A latch is provided at the control knob to lock carburetor heat off when not required. It is recommended that the control knob be unlatched (to activate carburetor heat assist) whenever OAT is between 80 degrees Fahrenheit (27 degrees Celsius) and 25 degrees Fahrenheit (-4 degrees Celsius) and the difference between dew point and OAT is less than 20 degrees Fahrenheit (11 degrees Celsius). Readjust carburetor heat as necessary following any change in power.

METEOROLOGICAL INFORMATION

The closest official weather observation station was Oceanside Municipal Airport, Oceanside, California (OKB), located 25 nautical miles (nm) southwest of the accident site. The elevation of the weather observation station was 28 feet msl. A routine aviation weather report (METAR) for OKB was issued at 1456 reported: sky conditions overcast at 2,000 feet; visibility 10 statute miles (sm); winds from 260 degrees at 3 knots; temperature 60 degrees Fahrenheit; dew point 52 degrees Fahrenheit; altimeter 30.02 inHg.

The METAR report issued at 1556 was identical to the 1456 METAR report, with the exception of winds, which were calm, and visibility was reported as 9 sm.

According to the Carburetor Icing Probability chart, the temperature and dew point spread indicated moderate icing was possible at cruise power or serious icing at glide power.

TESTS AND RESEARCH

The airframe and engine were inspected on January 28, 2004, at Aircraft Recovery Services, Littlerock, California. A National Transportation Safety Board investigator, along with manufacturers representatives from Textron Lycoming, and Robinson Helicopter Company, parties to the investigation, examined the helicopter.

Investigators examined the airframe and noted that the instrument panel was separated from its cockpit mount, and recovery personnel had cut the electrical wires. The clutch start circuit breaker had been tripped. For recovery purposes, the recovery crew had cut the tail boom into two sections. Investigators manually rotated both the main rotor and tail rotor gearboxes, and noted no binding. The main rotor mast tube (non-rotating) had fractured at the bottom side of the mast cowling. The fracture faces were jagged and displayed 45-degree shear lips. The airframe manufacturer established flight control continuity from the cockpit to the main rotor hub. Investigators noted that the tail rotor controls had been cut by recovery personnel; however, there were no abnormal breaks noted. The main rotor blades had been cut near the blade roots. Investigators noted that both main rotor blades were bent in an upward direction.

Investigators examined the engine and noted that it remained attached to the airframe by the engine mounts. There was no obvious external damage. The spark plugs were removed. The numbers 1 and 3 bottom spark plugs were coated in oil; however, the remaining spark plugs were gray in color, and the electrodes displayed an oval wear pattern. Magneto-to-engine timing was established by the engine manufacturer and was within the manufacturer's specifications. The engine manufacturer obtained compression in firing order by manually rotating the engine crankshaft by the cooling fan. Mechanical continuity was established throughout the engine and accessory sections. A borescope inspection of the cylinder combustion chambers showed no evidence of foreign object ingestion or detonation. No oil residue was observed in the exhaust system gas path. According to the engine manufacturer, the exhaust gas path and exhaust system components, the combustion chambers, and spark plugs all displayed coloration consistent with normal operation.

The engine manufacturer noted that the carburetor was undamaged and secured at its mounting flange. The throttle and mixture controls were securely attached at their respective control arms on the carburetor. The throttle and mixture controls were manually activated with no binding noted, thus establishing engine control continuity to the cockpit. The engine manufacturer further noted that the engine fuel lines were in place and secure at their respective fittings. While inspecting the fuel system prior to conducting an engine run, investigators found water in the gascolator and left fuel tank. It was discarded with no further discrepancies noted with the fuel system. Fuel was added to both fuel tanks and the engine was started utilizing the airframe manufacturer's standard starting procedures. Investigators ran the engine at various power settings with no anomalies noted. The magnetos were functionally checked during the engine run. The throttle was advanced to 1,600 rpm's, and the engine manufacturer selected each magneto individually. Investigators noted that the drop was within the manufacturer's specifications. There were no anomalies noted with the engine run.

NTSB Probable Cause

a loss of engine power for undetermined reasons.