NTSB Factual Report

HISTORY OF FLIGHT

On May 18, 2012, at 1212 Pacific standard time, a Hawker Hunter Mk 58, single-seat turbojet fighter aircraft, N329AX, operated by ATAC (Airborne Tactical Advantage Company) under contract to Naval Air Systems Command (NAVAIR) crashed while on approach to Naval Air Station Point Mugu, California (NTD). The sole pilot aboard was killed, and the airplane was destroyed by impact forces. The flight was conducted under the provisions of a contract between ATAC and the U.S. Navy to provide ATAC owned and operated aircraft to support adversary and electronic warfare training with VMFAT-101 (Marine Fighter Attack Training Squadron 101). The airplane was operating as a non-military public aircraft under the provisions of Title 49 of the United States Code Sections 40102 and 40125.

The airplane departed NTD at 1113 as the wingman in a flight of two Hunters, intending to participate in a fleet training exercise in off shore warning area W291. The flight's radio call sign was "COUGR21."

As the accident airplane was not equipped with any recording devices, and all radio communications with the accident pilot were via air-to-air with the flight lead, the sequence of events is based upon the flight lead's interview with investigators and information he separately provided to ATAC personnel following the accident.

According to the flight lead, shortly after takeoff, the accident pilot advised him, on an air-to-air frequency, that his lateral stick trim was almost all the way to the right in order to maintain wings level flight. The pilots discussed a maintenance issue involving a fuel transfer valve two days before the accident flight and concluded that an under-fueled right outboard tank was the likely cause.. The flight lead asked the accident pilot if he had checked the tank during preflight and if there was fuel in it and the accident pilot replied "I don't know." The flight lead said he recommended that the accident pilot return to the airport, "because we knew we had an issue right off the bat." The accident pilot elected to continue the flight and shortly afterward reported that the right outboard tank was indicating empty, much earlier than normally expected.

As the pilots continued to discuss the problem, the accident pilot also related that he believed that the airplane was also not drawing fuel from the left wing and external tanks. The accident pilot reported that he turned off the right boost pump, in an attempt to balance fuel, by configuring the fuel system so that the engine would draw fuel only from the left side.

After arrival in the training area, the flight lead indicated that he repeated his suggestion to return, but the accident pilot said he would give it more time to see if he could get fuel to transfer from the left side. Fuel transfer from the left outboard tank could only be confirmed once it empties (the magnetic indicator would change from black to white indicating empty). The accident pilot also reported that he saw the fuel quantity indicator on the left inboard tank began to show a reducing quantity in that tank, indicating 1300 pounds, and he thought he was getting a slow transfer from that tank.

Additionally, the low fuel light for the left feed tank illuminated, indicating that the tank was not full although there was still fuel in other tanks that should have transferred into the feed tank to keep it full. When this light illuminated, the accident pilot, at the flight lead's suggestion, cycled the left fuel transfer switch multiple times to read the front tank contents, as the gauge would indicate only the front tank quantity while the valve was in transit. At this point, after further discussions with the flight lead, the accident pilot decided to return to base because there were indications that fuel was not correctly transferring.

The flight lead elected to accompany him, and the pilots switched positions, with the accident pilot taking lead and flight lead flying echelon left (slightly above and behind to the left). During the return flight, the pilots continued to discuss the airplane systems and what indications the accident pilot should look for. The flight lead noted that the information relayed did not make sense regarding the fuel system compared to the cockpit indications. The accident pilot repeated that he thought he was getting a slow transfer from the left side fuel system, but could not determine the actual quantity of fuel in the left outboard tank (which would be the first to be depleted if fuel was transferring). He did not indicate the stick position or asymmetry. The pilots estimated how long it would normally take for the left outboard tank to reach empty, however, it was still indicating it contained fuel. About 10 to 15 minutes from the airport, the accident pilot related that he saw 1000 pounds remaining in the left inboard tank.

ATC restricted the flight to 15,000 feet until nearing the airport, and then cleared the flight for descent. Neither the pilots nor ATC declared an emergency or urgency situation. The flight lead stated that the accident pilot did not appear to have conducted a controllability check. The aircraft crossed the shoreline at about 6,000 to 7,000 feet on a left downwind leg for runway 21. They planned to extend the downwind leg to lose altitude and turn in for a long straight-in final approach. The flight lead reported that at a point approximately abeam the approach end of the runway at about 5,000 feet and an indicated airspeed of about 270 to 280 knots, the accident airplane began "a nice gentle left hand turn" which he said "didn't make sense." The flight lead said he was about 100 feet above and behind the accident airplane when it began to turn across his path, and it continued to turn and roll until it "turned into a barrel roll" on a course of about 030 degrees. He stated that it was a slow roll but it "was evident it was out of control." When the loss of control occurred, the flight had been airborne about 60 minutes.

The flight lead said that the accident airplane did not appear to slow down but he reactively maneuvered as needed to stay in position, although he noted that the airplane generally does not slow down noticeably in a descent if the throttle was already set at 6,000 rpm or less. They had not yet begun configuring for landing. The flight lead leveled off but banked left to observe the accident airplane pass below, which he said seemed to be briefly in level flight, and then "saw it snap over" and enter a steep nose down, left rolling maneuver. The flight lead reported he saw "lots of fuel coming out of the [external] tanks." He saw two objects come out as he was yelling "eject" on the radio, and initially thought they may have been the canopy and the ejection seat. He later concluded the objects were the left side external tanks separating from the wing. The airplane impacted less than three seconds later. He saw the two objects impact about 100 yards from the airplane (consistent with the ground position of the left external tanks). The airplane impacted at a steep angle and a small fireball was visible. The flight lead did not see a chute, and then he called the tower to launch the search and rescue effort.

Previous Flight of N329AX

On May 16, 2012, the airplane was flown by a different pilot on a similar mission, and experienced a fuel transfer incident. The accident pilot was the wingman on this flight. The aircraft fuel tanks had been filled and he also had an electronic warfare pod on the right wing station.

The preflight, start-up, and taxi out were uneventful. On initial climb-out, he started noticing an asymmetry developing so he checked the fuel system. He said the left transfer system indicator was showing no fuel was coming from the left internal or external tanks to the feeder tank. He cycled the selector switch between Auto, Rear, and Wing, six to eight times without any changes. During this time he kept adding right stick and right trim as no fuel was burning from the left wing tanks. At this point he was about 4 to 5 minutes into the flight when the left low fuel light came on indicating fuel was being drawn from the feeder tank. He elected to turn off the left fuel boost pump to preserve fuel in the left feeder tank.

The incident pilot said his wingman (the accident pilot) stated that nothing appeared unusual from the exterior and no fuel appeared to be venting. He applied a positive and negative G which did not have any affect. He then decided to return to land and his wingman continued the mission. During the return to the airport, he conducted a controllability check where he found that the left wing was heavier than the right wing. About that time he saw that the right wing outboard tank magnetic indicator turned white, indicating that tank was empty. He landed on runway 3 with no difficulty, and estimated the total flight time as about 17 to 18 minutes.

Troubleshooting and Maintenance

After landing, and before the incident pilot shut down the engine, maintenance personnel checked the fuel transfer valve and motor while he cycled the left wing switch to rear but the transfer valve was not working. The attempt to troubleshoot took about 15 minutes. (Tests at Lortie Aviation indicated the engine ground idle fuel burn was about 145 pounds per hour.) He believed that after shutdown, the left side exterior tanks were still full, the right outboard was empty, and the right inboard had burned some fuel but was not yet empty. Investigators estimated that the fuel burn during the incident flight would have left the airplane with a shutdown fuel load of approximately 6000 pounds. The incident pilot conducted a post-flight debrief with the wingman (accident pilot) after his return.

The ATAC Point Mugu Maintenance Manager (MM) took part in the troubleshooting. He said the procedure they followed was to first open the access panel on the left underside of the fuselage, near the gear well, to access the valve. They then manually selected the fuel transfer switch from wing to rear (which should have activated a motor that moves a ball valve), and heard no left motor sound, and saw no movement of the valve or indicator. They performed the same task to the right side, and all indications sounded and looked normal. They then disconnected the cannon plug from the motor to check the wiring. No power was measured at the motor, and the associated 5 amp fuse was found to be blown. The shop did not have any spare 5 amp fuses, so installed a 7.5 amp fuse to test, and reconnected the cannon plug and the fuse blew right away, with no noise from motor. Actuating the switch with the motor cannon plug disconnected did not cause a blown fuse. The MM reported the problem to Lortie Aviation via a web based system for requesting spare parts and other maintenance information. Lortie overnight-shipped a replacement motor to ATAC.

The following day, the replacement part arrived about 1000. The ATAC mechanic assigned to the job was working the afternoon shift and began preparing for the replacement upon arrival. Neither he nor other ATAC personnel at the Point Mugu base, had ever done this particular task before. The mechanic had participated in the troubleshooting the day prior. He stated that the job took "all day" because the location of the motor and valve was very difficult to access. He first removed the motor and gave it to the MM for bench testing while he began to install the replacement motor. The MM applied power on a test bench, with the same result as on-wing. The valve shaft could not be accessed by hand with the motor and lines installed, during the time that the motor was removed, none of the maintenance personnel tested for motion of the ball valve.

The mechanic said that due to the difficult access he had to use a crows foot and extensions on his wrench to attach the motor to the valve. The mechanic described that the motor attached to the valve case by two threaded studs that protrude from the motor fastened with nuts. The studs and the electrical cannon plug are aligned so that the motor body can only be installed one way. He described that the valve shaft protrudes and has a slot in the end of the shaft, which fit into a receptacle on the motor which contained a cross-pin that fit into the slot. He reported that when he first tried to slide the motor over the shaft, it would not fit, because the shaft was not aligned. He removed the motor and turned the shaft "not as much as a quarter turn, maybe a sixteenth." He said that he used a standard screwdriver to turn the valve shaft, but it was not difficult to turn. He reported that he could not see the face of the valve housing or any markings thereon.

He replaced the cannon plug and they performed a check of the system. The fuse was replaced and the MM cycled the transfer switch. The mechanic could hear the movement at the valve, and feel vibration by hand. The indicator on the motor casing was not visible with all lines installed. The MM observed normal indications in the cockpit during the check. It was not possible to test the AUTO function without getting into the aft fuselage tanks. The mechanic then replaced the duct work that had been removed for access and secured the area. Another mechanic looked over the area for general appearance and absence of debris. They left the access hatches open so the inspector could sign off on the repair the next morning.

ATAC Hunter maintenance training was primarily conducted using on the job training methods. There was no formal school or training location for the airplane, although the maintainer who replaced the motor was an FAA A&P certified mechanic. Maintenance tasks were referenced to the British airplane manual, and ATAC had access to Lortie technical representatives by phone to give guidance and answer questions about specific tasks.

When the mechanic came in the next day, Friday May 18, the pilots were already there. He assisted with the daily checks of other airplanes, and the two pilots departed with other airplanes for the first sortie. The inspector signed off on the motor replacement, but the hatch would not close properly, the mechanic repositioned the duct and closed up the hatches, then took the airplane out of the hangar to the ramp for refueling.

The mechanic that performed the motor replacement also did the refueling. He described to investigators that he cycled the fuel tank indicator selector through the positions to determine which tanks needed fueling (see Aircraft Information for a description of the refueling panel operation). He described fueling the aft fuselage tanks first, then noted the inboard external tanks indicated full, he switched the indicator selector again for the outboard external tanks and described that he "went to tap on the tanks and see how much is in there, if I can hear it." After going to the right outboard, then left outboard, he returned to the refueling panel to reset the battery timer rotary switch ("egg timer") and said he saw all the lights were extinguished indicating the tanks were full, so he stopped fueling. The mechanic described that on a recent occasion, a tank overflowed and he did not want to risk that. He estimated a total of 275 or 277 gallons was added, on top of the remainder from the early return flight. He assisted the accident pilot with the preflight inspection, but could not recall if he or the pilot tapped the external tanks.

INJURIES TO PERSONS

The pilot was fatally injured.

DAMAGE TO AIRPLANE

The airplane was destroyed by impact forces. The airplane was highly fragmented and mostly contained in an impact crater. See wreckage and impact information.

PERSONNEL INFORMATION

The pilot, age 57, had worked for ATAC since February of 2011. He was a former U.S. Navy bombardier-navigator in the A-6 Intruder before transitioning to pilot and flying EA-6 Prowlers and F/A-18 Hornets. He served as a Carrier Air Wing Commander, and Executive Officer on the USS Kitty Hawk CVN-63. He reported 8,325 hours total flight time, and held an FAA Airline Transport Pilot Certificate with type ratings in the Gulfstream

NTSB Probable Cause

the pilot's decision to continue the flight with a known fuel imbalance condition that resulted in a loss of lateral control when the imbalance exceeded the known capabilities of the airplane. The fuel imbalance was due to incomplete refueling and an ineffective preflight inspection by the pilot. The imbalance was further complicated by an incorrectly assembled fuel transfer valve and motor combination.

Contributing to the severity of the accident was the pilot's delayed decision to eject prior to exceeding the ejection seat envelope. Also contributing to the accident was (1) the Navy's oversight environment, which did not require airman, aircraft, and risk management controls or standards expected of a commercial civil aviation operation, and (2) ATAC's organizational environment, which did not include CRM training to promote good aeronautical decision-making and ORM guidance to mitigate hazards. Also contributing to the accident were the design features of the airplane, which were typical of its generation, including the lack of accurate fuel quantity indications, the design of the fuel transfer valve; and the maintenance program's lack of clearly documented procedures and type-specific training for the Hunter.