text
stringlengths
0
3.21k
The fuel cell cover was opened and water, with no evidence of fuel, was found inside the cell.
Potential water entry points included the fuel vent system, a deficient fuel-cap seal, or tank damage.
The investigation considered the possibility that damage, including torn/ripped structure around the fuel cell, might have compromised the fuel cell.
However, no evidence of fuel leakage was observed, and no visible holes or tears were noted in the fuel bladder.
The contents drained from the fuel pump filter bowl contained significant amounts of entrained particulates, and a liquid suspected to be water.
Water-detecting paste confirmed the presence of water.
The filter bowl in the housing of the engine-driven fuel pump was opened and examined.
The filter exhibited significant contamination of unknown particulate matter, a paste-like substance, and what appeared to be plant material.
With low or zero fuel pressure, such as when the engine is not operating, the fuel supply line from the fuel pump filter to the FSN is normally closed at both ends by spring-loaded check valves.
These check valves trap the fluid immediately prior to its introduction into the FSN and combustion chamber.
The fluid from that line was drained and examined.
That fluid was observed to be about 30% water.
The fuel system architecture precluded introduction of water into either the fuel pump or the FSN fuel line unless the engine was operating.
Maintenance Records Hansen personnel reported that a mechanic was stationed on the fishing boat with the helicopter, but they provided only his name to the investigation.
They did not state whether he was a certificated mechanic, and did not provide any mechanic certificate or qualifications information for him.
In addition, Hansen reported that the mechanic was sent home to the Philippines after the accident.
The investigation did not attempt to contact the mechanic.
The Hansen Director of Maintenance (DOM) provided the investigation with a binder that he represented as being the helicopter maintenance records.
The binder included a mix of flight records, status reports, and copies of FAA 337 forms.
Exclusive of the 337 forms, none of the contents conformed to the FAA maintenance entry requirements.
The records contained multiple internal service time and/or component number discrepancies.
According to the FAA inspector, cursory comparisons of the 337 forms with the records on file with the FAA in Oklahoma City revealed numerous discrepancies.
The most recent recorded 100 hour/Annual, 300 hour, or 600 hour inspection was completed and signed off by the Hansen Helicopters DOM on 5/7/16.
On that inspection entry, the airframe time was listed as 6,891.1 hours, and the "Hobbs time" was listed as 544.1 hours.
The inspection entry stated "Next inspection due is a 100 hour at 6991.1" [hours].
However, despite the fact that all available information indicated that the helicopter has accumulated nearly 400 hours since that inspection, no additional FAA-compliant inspection entries were observed for dates subsequent to 5/17/16.
Fuel Contamination Source After the damaged helicopter was delivered to Guam by the fishing boat, the boat departed without examination by either Hansen Helicopters personnel or any investigation-related personnel.
Therefore, the specifics of the boat's storage and dispensing system for the helicopter fuel, or the equipment and procedures related to prevention and detection of fuel contamination, were not able to be determined.
Thus, the investigation was unable to determine the fishing boat's potential as the source of the water contamination of the fuel.
Additionally, because the investigation was unable to interview the pilot, the specifics of his activities and procedures, including preflight inspection, regarding the prevention or detection of fuel contamination, were not able to be determined.
Although the evidence was consistent with the water being present in the helicopter fuel system prior to the flight, the investigation was unable to determine when or how the water entered the fuel system, or why the pilot failed to detect the water in the fuel.
Helicopter Fuel System Information The helicopter was equipped with two fuel cells that were interconnected.
The two fuel cells were of the conventional bladder type, and were located under the cabin floor in separate compartments.
The helicopter was equipped with a single fuel filler neck and cap, located on the right side of the helicopter, aft of the cabin door.
The fuel cell sump drain was in the left fuel cell.
There was one drain valve located on the lower fuselage in this sump area.
It was spring-loaded to the closed position, and depressed (pushed in) to open.
In addition, the helicopter was equipped with a drain valve that installed on the fuel line elbow assembly that was attached to the engine firewall.
Other Helicopter Systems and Flight Procedures Helicopter electrical power was provided by a 24 volt battery, and a 28 volt starter-generator that was gear-driven by the engine.
Generator output was controlled by a voltage regulator.
The helicopter was equipped with a visual and aural caution/warning alerting system, part of which was an array of discrete, dedicated lights across the top of the instrument console.
That alerting system was unable to be activated or tested during the examination due to a lack of electrical power on the helicopter.
One of those discrete annunciator lights was the "GEN OUT" light.
According to the helicopter manufacturer's information, the dedicated "GEN OUT" annunciator light in the caution/warning array will illuminate when the generator "is not powering the electrical bus." A loss of engine power would result in such a condition and GEN OUT alert, among others.
According to the helicopter manufacturer's guidance in the Rotorcraft Flight Manual (RFM), in the event of a generator failure, the pilot is to "Turn generator switch off" and "Reduce electrical load to 16 amperes or less, if possible." The helicopter was equipped with N1, N2, and [Main] Rotor rpm gauges, also referred to as "tachometers." The N2 and Rotor rpm indications were presented on a single instrument, with two concentric scales and two indicating needles, one for each parameter.
The scales were calibrated and positioned so that during normal operation, the N2 and Rotor rpm needles will be aligned with one another.
The tachometers were marked with green arcs and red radial lines to respectively denote normal operating range and minimum and maximum rpm values.
The N2 scale was from 0 to 120%, and occupied an arc of about 290掳.
The N2 lower and upper rpm values were 100% and 103% respectively, and therefore occupied an arc of about 8掳.
Engine power loss in this model helicopter will typically initially manifest itself with left yaw, decreases in engine and rotor rpm, and a change in noise level.
Subsequent manifestations will include airspeed and altitude losses.
To assist pilot detection of an engine failure, some helicopters of this model were equipped with an "Engine Out" alerting system.
The system included a dedicated "ENG OUT" annunciator light, augmented by an aural warning horn.
The generator switch must be ON to enable the Engine Out warning.
N1 decrease below 55% will trigger these ENG OUT alert annunciations.
The accident helicopter was equipped with the ENG OUT annunciator light.
However, the investigation was unable to determine the presence or condition of any of the other components of the Engine Out alerting system, or the pre- or post-accident functionality of that system, if in fact it was installed and intact.
The RFM procedures for an engine failure when operating more than 420 ft above the surface specified that the pilot should "enter normal autorotation by lower[ing the] collective pitch full down" and then selecting an appropriate landing spot and airspeed.
The RFM also stated that an engine restart can be attempted at the pilot's "discretion." For engine failures at altitudes below 420 ft, the RFM specified lowering of the collective to maintain minimum rotor rpm, and stated that the "amount and duration of collective reduction depends upon the height above the ground at which the engine failure occurs." The pilot did not provide any indication that he was alerted to or noticed an engine power loss until after he became involved in addressing the generator problem, and the investigation was unable to question the pilot on his observations or actions.
Ownership and Control The accident report filed by Hansen Helicopters stated that the form had been completed by the pilot, and that "Jim's Air Repair" was the operator of the helicopter.
A Hansen representative stated that two organizations were "affiliated companies," but did not provide any additional details at that time.
In an email dated March 14, 2017, the representative stated that "Hansen Helicopters provides employment recruiting, training and logistical support for Jims Air Repair." In that same email, the Hansen representative also stated that the pilot was "working under a contractor's agreement with Jim's Air Repair out of Vanuatu." Hansen Helicopters' primary facilities were in the US state of Georgia, and on Guam, a US territory.
NTSB requests for documentation regarding the operational arrangements between Hansen Helicopters, Jim's Air Repair, the pilot, the maintenance providers, and the Japanese fishing boat were not satisfied; therefore the investigation was unable to independently determine which personnel and companies exercised the actual operational and maintenance control of the helicopter.
Injury Reporting Accuracy The Hansen-filed written accident report to the NTSB indicated that the two persons on board sustained minor injuries, and Hansen never advised the NTSB of any changes to that status.
About 13 days after the accident, the NTSB was advised via a third party that both the pilot and the observer had been hospitalized since the accident, as a result of injuries incurred in the accident.
The NTSB was further advised by this third party that the observer had already been transferred to Japan, and that the pilot was scheduled to be transported to the Philippines for surgery for injuries sustained in the accident.
The investigation was able to confirm that the pilot was seriously injured, but was unable to confirm the level or nature of the observer's injuries.
Before the accident flight, the airplane had sat in its hangar for the previous 2 months with its fuel tanks half full under varying temperature conditions.
The pilot had planned on flying to a safety seminar that began at 1630, so he had the airplane pulled out of its hangar, and its main fuel tanks were topped off from a fuel truck.
After his arrival at the airport shortly before 1700, the pilot performed a preflight inspection.
The manager of the hangar facility described the pilot's preflight inspection as "real quick." A lineman observed the pilot in a position to reach the fuel strainer valve, but he did not see the pilot sumping the main fuel tanks.
When the lineman drove by the airplane, he saw a puddle about 1 foot in diameter on the tarmac beneath the fuel strainer, but he did not note anything under either main fuel tank drain.
The lineman also noted that the airplane had an underinflated tire, but, due to other duties, he could not warn the pilot before he taxied the airplane away.
About 2 minutes after takeoff, the pilot reported an "engine problem" to air traffic control and turned the airplane back toward the airport.
The airplane subsequently descended at a steep angle, consistent with a stall, into a house located in a populated area.
The airplane impacted the roof, came to rest upside down, and was subsequently mostly consumed in a postcrash fire.
Postaccident examination of the airplane revealed evidence indicating that the airplane was not under power at the time of the accident; the fuel-injected engine was charred and the propeller did not exhibit torsional bending or leading edge damage that would have been present if it had been under power.
No preexisting mechanical anomalies were found that would have precluded normal engine operation.
However, when the fuel flow divider was opened, water was found in it, which likely resulted in the loss of engine power.
Water typically enters fuel tanks via three sources: leakage, normally through a fuel cap; contaminated fuel sources; and fuel tank condensation.
The fuel cap was likely not the source of water since the airplane was stored in a hangar.
Contaminated fuel from the fuel truck was also not the likely source of water since the truck was reportedly sumped daily.
Further, on the day of the accident, five airplanes received fuel from the same truck before the accident airplane with no reports of any performance anomalies, and a clean fuel sample was taken from the truck about 20 minutes after the accident.
It is more likely that condensation occurred in the half-filled fuel tanks during the previous 2 months that the airplane was sitting in the hangar under varying temperature conditions.
Regardless, the pilot had an opportunity to eliminate the condensation during the preflight inspection.
However, as noted previously, not enough evidence existed to determine whether the pilot actually drained each main tank to ensure that all of the water was removed.
It is likely that the pilot either did not sufficiently drain the main fuel tanks or that he was relying on draining the main fuel tanks through the fuel strainer and fuel lines and did not sufficiently drain them all.
Given witness statements indicating that the pilot was in a hurry and his oversight of the underinflated tire, it is likely that the pilot's preflight inspection was inadequate, which resulted in his failure to notice the fuel tank condensation.
The pilot's inadequate preflight inspection, which resulted in his failure to note the water in the fuel tank due to condensation, which subsequently shut down the engine in flight.
Contributing to the accident was the pilot's self-induced pressure to expedite the departure.
On November 13, 2012, about 1715 central standard time, a Piper PA-32-300, N717RL, was substantially damaged when it impacted a house in Jackson, Mississippi.
The airline transport pilot (ATP) and the two pilot-rated passengers were fatally injured.
Visual meteorological conditions prevailed, and no flight plan was filed for the flight, from Hawkins Field (HKS), Jackson, Mississippi, to John Bell Williams Airport (JVW), Raymond, Mississippi.
The personal flight was conducted under the provisions of 14 Code of Federal Regulations Part 91.
According to the owner of the company to which the airplane was registered, and who was also a student pilot of the ATP, the ATP and he were going to fly to JVW to attend a Federal Aviation Administration (FAA) safety seminar.
The owner was subsequently unable to go, but advised the ATP that the airplane needed to be flown since it hadn't flown since Labor Day [September 3rd].
The owner was unaware that the other two pilots were onboard.
Announcements for the safety seminar indicated that it was scheduled to begin at 1630.
According to the manager and a lineman at a local fixed base operator (FBO), the airplane was pulled out of its hangar and the main fuel tanks were topped off prior to the arrival of the ATP and the passengers, shortly before 1700.
Both indicated that the ATP's preflight inspection was much quicker than normal.