NTSB Factual Report

HISTORY OF FLIGHT

On July 11, 2008, at 1050 eastern daylight time, an Agusta A109E, N109DU, operated by AirMed, Incorporated, sustained minor damage following a forced landing after a partial loss of engine power after takeoff from Doctors Hospital Heliport (8GA4), Augusta, Georgia. The certificated commercial pilot and the two crew members were not injured. Visual meteorological conditions prevailed for the flight that originated from 8GA4 and was destined for Brevard, North Carolina. A company flight plan was filed for the positioning flight that was conducted under the provisions of 14 Code of Federal Regulations Part 91.

In a written statement, the pilot described ascending to a hover over the helipad, conducting a power check, and determining that the helicopter was "ready for flight." The helicopter departed in a northwesterly direction, climbed to an altitude of about 30 feet, where "both engines lost power." The pilot performed a right pedal turn and maneuvered the helicopter back toward the helipad. The helicopter landed short of the helipad, straddling a fence that separated the pad from an adjacent road. The helicopter's main landing gear sustained minor damage, and the helicopter's skin was punctured in a non-structural area.

PERSONNEL INFORMATION

The pilot held a commercial pilot certificate with ratings for airplane single engine land, rotorcraft helicopter, as well as instrument airplane and helicopter. His most recent FAA second class medical certificate was issued on October 24, 2007. The pilot reported 6,400 total hours of flight experience, 6,000 hours of which were in rotorcraft, 950 hours of which were in the Agusta 109 line of helicopters, and 18 hours of experience in the Turbomeca-powered Agusta A109E. The pilot reported that all 18 hours were in the accident helicopter.

AIRCRAFT INFORMATION

According to the National Transportation Safety Board Form 6120.1, the helicopter had accrued 2,800 total aircraft hours. The helicopter was on a manufacturer's continuous inspection program, and its most recent inspection was completed on April 17, 2008.

METEOROLOGICAL INFORMATION

At 1053, the weather reported at Augusta Regional Airport (AGS), 8 miles southwest of the site, included scattered clouds at 2,300 feet and winds from 310 degrees at 4 knots. The visibility was 10 miles. The temperature was 28 degrees Celsius (C) and the dew point was 23 degrees C.

TESTS AND RESEARCH

The digital engine control units (DECU) for both the number 1 and number 2 engines were removed from the helicopter, shipped to Turbomeca USA, Grand Prairie, Texas for download and examination under the supervision and direction of a Federal Aviation Administration (FAA) aviation safety inspector. Examination of the data revealed a "Stepper Motor Channel Fault" on the number 2 engine. Both DECUs were then returned to the helicopter for reinstallation and further troubleshooting on the airframe.

Both DECUs were reinstalled in their original positions on the helicopter, electrical power was applied, and no faults were immediately noted. Power was removed from the helicopter and then reapplied, which produced an "Engine Stepper Motor Channel Fault" on the number 2 engine. The fault annunciated an "ECU FAIL" segment light, illuminated the "Master Warning Lights," and produced the applicable codes on the Engine Display Unit (EDU) Maintenance Page. The audio warning system produced an aural warning as well. This step was repeated for a total of seven power applications with identical results.

The DECUs were reinstalled, however, in the opposite position from the original test to determine if the DECU was the source of the fault. The helicopter was then powered up multiple times, and consistently produced the same stepper motor fault on the number 2 engine. The number 2 hydro-mechanical unit (HMU) was then removed and replaced. Once power was applied, there were no faults noted on either engine. The DECUs were then swapped back to their original positions, power was applied, and no faults were recorded.

On October 13, 2008, the number 2 DECU, the number 2 HMU, and the wire harnesses that connected them were tested at Aquitaine Electronique and Turbomeca, France under the supervision of Bureau d'Enquêtes et d'Analyses (BEA) France. The DECU and the HMU both functioned as designed with no faults noted.

Two harnesses (Control and Monitoring – Regulation) were tested and impedance was detected on pin 3 of the stepper motor plug on the regulation harness. The jacket of the harness was removed, which revealed "moist deposits" and corrosion on the metallic ground shield. Removal of the shield revealed more deposits and corrosion, as well as cracked and burned insulation on wire 3.

Further examination of the data downloaded from the DECU's revealed that the stepper motor channel fault on the number 2 DECU was recorded 98 seconds after power-up with the engine mode switch in the idle position. The fault commanded the number 2 HMU fuel metering needle to freeze at 65 percent N1. Further control of engine speed required activation of the number 2 engine trim switch on the collective control stick, or a direct input on the number 2 engine power lever. Neither action was taken during the accident flight.

Examination of data from the number 1 DECU revealed a level 1 “collective pitch channel fault” 209 seconds after DECU power-up with the number 1 engine mode switch in the “FLIGHT” position. Detection of this level of fault (level 1) has no effect on the control system and is not displayed to the pilot in flight. The parameters detected at the time of the fault showed that the DECU torque limited the engine according to its control law, and the “limit override button” in the “OFF” position. Activation of the limit override button by the pilot would have allowed the DECU to increase that limit. The data also showed that the “102% NR” switch was not activated at any point during the accident flight, as required by the flight manual.

According to the Agusta A109E Rotorcraft Flight Manual, the 7th and 8th items in the “Before Takeoff” checklist are:

RPM switch : Set 102%

NR/N2 : 102% stabilized, check.

ADDITIONAL INFORMATION

In a subsequent statement, the pilot questioned how there was no indication of one-engine-inoperative (OEI) time recorded on the number one DECU during the incident flight, had the number two engine not accelerated past the idle position.

As a result, the data was re-examined, and a 17-second OEI event was identified previous to the accident flight. Review of maintenance records confirmed that the event took place on May 7, 2007.

Closer examination of the memory portion of the DECU download, column three, showed “System condition-word 1” at the time of the "Collective pitch channel failure" fault on the #1 engine during the incident sequence. This indicated the engine was in the 2.5 minute OEI range as shown by the parameter titled “Use 2.5 min OEI rating = ON.” According to the engine manufacturer, the OEI timer does not accrue time until the event exceeds 10 seconds.

According to the operator's director of maintenance, the helicopter was not equipped with an optional "Engine Out Kit" that was offered by the manufacturer of the helicopter. According to the manufacturer's technical bulletin (109EP-81)," With the subject kit installed, the pilot receives in the headset the “ENGINE OUT” audio signal in case, during the take off phase, one or both the engines are in IDLE or in OFF.

NTSB Probable Cause

The lack of available engine power and subsequent loss of rotor rpm due to the pilot’s departing with one engine at idle. Contributing to the incident was a corroded engine control wiring harness.