C-130 Broad Area Review (2023)

TABLE OF CONTENTS

Tasking From The Secretary of the Air Force

The report you are about to read answers the Secretary of the Air Force’sdirection to conduct a BAR of C-130 flight safety. Our goal was to make it understandable,by explaining or eliminating as much service jargon as possible. It represents ourteam’s review of C-130 aircraft systems, training programs, maintenance activities,and flying operations.

The Secretary of the Air Force, responding to a request from members of the UnitedStates Senate, tasked the Acting Chief of Staff of the United States Air Force to conducta BAR of C-130 flight safety, addressing the missions and environments in which it flies,and its safety history. In addition, the Secretary directed the team to look at the King56 incident to make certain all potential causes and contributing factors were properlyconsidered, and to ensure that everything appropriate was being done to enhance C-130flight safety.

Executive Summary

Team Composition

In response to the Secretary of the Air Force’s (SECAF) direction to form theteam, the team chief selected a group of individuals with a broad background of experiencein the C-130 weapon system. Representing a combined total of 45,000 flight hours and over235 man-years experience with the aircraft, the team included operators and maintainersfrom the using commands (including the Air National Guard and Air Force Reserve Command),Air Force Safety Center, Air Force Materiel Command and Headquarters Air Mobility Command.There were members from logistics, operations, and the National Transportation SafetyBoard. Representatives of the aircraft’s manufacturer and major component suppliers(i.e., engine and propeller manufacturers) served as advisors to the team.

Approach

Our approach was simple and direct--get the right experts, get the facts (firsthandwhenever possible, from the field operators and maintainers), determine safety issueswithin C-130 operations and maintenance (including training and aircraft systems, andaircraft safety data) identify possible scenarios for King 56, evaluate each, then report.To execute that approach, the BAR traveled to sixteen C-130 operating and maintenancelocations across the country (Little Rock AFB, AR; Pope AFB, NC; Duke Field, FL; HurlburtField, FL, Moody AFB, GA; Keesler AFB, MS; Harrisburg ANG, PA; Warner-Robins ALC, GA;Kirtland AFB, NM; Davis-Monthan AFB, AZ; Youngstown ANG, OH; Dobbins AFB, GA;Lockheed-Martin Corporation, Marietta, GA; Moffett Field, CA; Schenectady, NY; andPortland ANG, OR), spoke with literally hundreds of crew members and maintainers,performed and analyzed tests, and read reports. The BAR also took leads from a toll freenumber that the BAR established to help us gather the information the BAR needed.

Fleet Safety Record

The team came away from this review convinced that the C-130 has been, and remains, avery safe and dependable aircraft. The team became increasingly aware of crew membersuspicions that the aircraft’s synchrophaser (an electronic device used tosynchronize the propellers and phase the passing of their blades as they turn so as toreduce propeller-induced noise and vibration), was to blame for King 56 and other enginepower loss incidents. The team found that the internal failure of the synchrophaser wasrarely the cause of the incidents examined (2 of 71). When involved, it turned out to beonly a contributory factor to the problem, usually as a result of faulty signals fed fromother systems, or by electromagnetic interference (EMI).

The BAR examined the C-130 fleet safety record and found that, in almost 25 millionflight hours worldwide, there had not been a recorded instance before the King 56 accidentwhere all four engines quit running in flight. The team reviewed data from multiplesources (including: contractor manuals, aircraft technical orders, safety and accidentreports, and aircrew and maintenance interviews) and analyzed the information, as well asaircraft test results, at considerable length.

In the case of King 56’s uncommanded power-loss, the team believes the most likelyexplanation remains fuel starvation, due to one of several possible causes, each of whichthe BAR has evaluated in detail within the body of the report. Although the BAR believesit was fuel-related, no single specific cause could be conclusively determined. In thisreport, the BAR presents the scenarios that explain how this loss of fuel to the enginesmight have happened. The facts the BAR established, combined with test results referencedunder each scenario, enabled the BAR to narrow the focus to four likely scenarios.

Action Taken

The Air Force published a safety supplement to change the flight manual emergencyprocedures for dealing with four-engine or multiple engine power-loss. Prior to publishingthis report, the team drafted this bold face procedure which crews must commit to memoryand use immediately in such an emergency to improve their odds of solving the problembefore it becomes unrecoverable. The crews exposed to this procedure overwhelminglyapproved of it and were quick to provide feedback on how to improve upon it.

Recommendations

The BAR’s recommendations, listed below, are divided into three categories:general, C-130 specific, and King 56 salvage.

A. General Recommendations:

1. Lead Command Operating Instruction: The Air Force should review and updatethe existing lead command operating instruction to:

a. Fully reflect changes which have occurred since the CONUS theater airlift fleettransferred from Air Combat Command to Air Mobility Command.

b. More fully define the lead command’s leadership role and its responsibilities,particularly with respect to configuration control (making certain that cockpitinstrumentation and aircraft modifications are standardized across a fleet of likeaircraft to facilitate standard operating and maintenance procedures). This leadershiprole should extend to cover generic operational issues as well.

c. Better define the lead command’s authority to enforce configuration control andthe accountability of other commands to the lead’s direction.

d. Empower the lead command and properly resource the lead and other user/supportingcommands to enable them to:

1. Update, consolidate and standardize aircraft flight manuals and operatingguidance to assure crews have current procedures and performance data.

2. Do the same for maintenance manuals to assure maintainers have the up-to-dateinformation they need to properly maintain the aircraft.

2. Air Force, Federal Aviation Administration, and National Transportation SafetyBoard Standardized Flight Data Recorder Parameters: DFDR performance limitationsseverely hampered the King 56 investigations and this review. The Air Force shouldconsider the Federal Aviation Administration and the National Transportation Safety Boardguidelines and experience in arriving at a standardized set of digital flight datarecorder flight parameters. This would ensure that essential flight data is captured forevaluation in future incidents and accidents.

3. Ditching & Bailout Procedures: The Air Force should review ditching andbailout procedures. As part of this effort, the Air Force should:

a. Conduct an analysis of world-wide ditching events. That data should be used toupdate and standardize all flight manuals with an accurate discussion of ditchingsurvivability and techniques.

b. Review the information concerning bailout in the flight manuals for consistencybetween models of the same aircraft, and revalidate the accuracy of the informationprovided to the crews.

c. Establish a requirement for crews to review these procedures on the first leg ofeach over-water mission, in order to maintain reasonable familiarity with theseprocedures.

d. Establish a standard life support equipment requirement, appropriate for theaircraft’s missions, for each mission design series C-130 and equip each for thatrequirement.

B. C-130 Specific Recommendations

The team made the following recommendations based upon its BAR of C-130 missions,operating environments, and the fleet’s flight safety record (with particularemphasis placed on the 71 incidents the BAR examined involving uncommanded powerreduction):

1. C-130 Technical Orders: A total of 487 of the Air Force’s 627 C-130technical orders currently have an inordinate number of supplemental page inserts and arein need of a complete rewrite to incorporate the new information into the body of thetext. For several years there has not been sufficient funds available to complete therewrites. This important issue is broader than just the C-130 alone and is under reviewAir Force-wide. It will require a significant investment, over $20 million andapproximately two years to fix the C-130 alone, using current manpower levels to correct.The Air Force should fully fund this action, as well as new initiatives underway toconvert USAF technical manuals from the old, expensive and time-consuming paper format tothe newer digital format. New CD-ROM technology offers many benefits, including areduction in the annual $2.5 million cost of maintaining our T.O.s. This conversion facesmany obstacles, including the cost of conversion as well as training and equipping fieldunits to handle electronic data rather than paper.

2. EC-130 Commando Solo II Mission Evaluation: Until replaced with newer, morecapable C-130s, the Air Force should reevaluate and closely monitor the EC-130 CommandoSolo II mission.

C. King 56 Salvage

The BAR recommends the Air Force recover selected wreckage from King 56. The componentsof greatest interest are: the wing section, the fuselage tanks, and the cockpit fuelquantity gauges. These items could answer many open questions and provide additionalinformation concerning the various fuel-related scenarios. While the exact cause of theKing 56 mishap may never be known with absolute certainty, this wreckage could reveal aprobable cause and refute many scenarios. The most compelling reason to obtain additionalwreckage is the possibility that evidence might be found which points to an unknown newscenario.

Section 1.0

C-130 Operating Environments and Missions

1.1 Background

1.1.1 Since its introduction into the Air Force inventory in 1955, the C-130 has servedin a variety of operating environments and missions. From the polar regions to thetropics, this aircraft has delivered personnel, equipment, and supplies by a variety ofmeans to locations all over the planet. Over 2,000 aircraft support the United States andits Allies’ military operations, as well as numerous commercial operations.

1.1.2 There are few environments this aircraft has not operated in. It has served as alaunch platform for remotely piloted vehicles, and as a recovery platform for bothpersonnel and data packages. It routinely penetrates hurricanes, delivers ordnance,provides combat communications links, facilitates rescues on land or at sea, services ourremote stations at the North and South Pole, refuels aircraft, and broadcasts radio andtelevision messages when the mission requires. In the late 1950s, it provided a good dealof aerial photography for cartography used to this day in the United States. It has beenan air ambulance, and a deliverer of relief supplies to refugees, and a transporter ofrefugees to safety around the world. It has fought forest fires from California toIndonesia. By far, it’s most often seen as a theater airlifter, either air droppingor air landing troops, equipment, and supplies to wherever they are needed.

1.1.3 The aircraft used in this theater airlift role typically operate in the lowaltitude regime, flying a few hundred feet above the ground with their crews navigating bya combination of pilotage, self-contained navigation systems, global positioning systems,and "dead reckoning." Flying a series of carefully developed course lines,designed to avoid known threats while ingressing to their target drop or landing zones,these aircraft typically stay low until their destination, rising only to drop theirparatroops or pallets, or coming in for an "assault" (short field, i.e., 3,000foot long airstrip) landing. They exit the same way. Equipped in some cases with AdverseWeather Aerial Delivery Systems (AWADS) or their equivalent, they may drop their loadswithout ever seeing their target visually, but with impressive and reliable accuracy.Traveling singly or in formation, in daylight or on night vision goggles in blacked-outconfiguration, they deliver the goods where needed.

1.1.4 While the low-level portion of the flight exposes them to hazards from birdstrikes to small arms, anti-aircraft artillery fire, and Surface to Air Missiles (SAMs),it is the pass over the drop zone, or the time spent getting into and out of the assaultstrip, that is probably the most dangerous for the aircraft and crew. Relatively high inthe air and slow at that point, here they are most vulnerable to ground fire. A largeportion of the C-130 force trains for this kind of operation daily.

1.1.5 Not always used in the combat delivery mode, the basic airlift version of theC-130 has an intercontinental range, allowing it to carry a number of pallets of cargo, orup to 92 passengers. This range, and the aircraft’s versatility, made it a logicalcandidate for a number of modifications to support a variety of special missions.

1.2 C-130 Operating Environments and Missions

1.2.1 The C-130 is arguably the most versatile aircraft in the Air Forceinventory. Currently 52 units in the active force, Air National Guard and the Air ForceReserve Command fly the aircraft in the combat aerial delivery mode alone.

1.2.2 Combat Aerial Delivery: The most common mission is called combat aerialdelivery, or "CAD" for short. This term refers to delivering cargo by landing atan airfield (called "airland") or dropping it by parachute (called "aerialdelivery"). The airland mission involves operating the aircraft into airfieldsworldwide, from large, busy commercial airports like Chicago, O’Hare, to small,isolated, unimproved dirt landing zones bulldozed and hacked out of almost anyplace, whichcan be as short as 3,000 feet long. The aerial delivery element of CAD involves airdropping personnel and equipment following ingress. This can be done as a single aircraftor as part of a large formation. Airplanes get to the drop zone by making use of eithervisual procedures or in Instrument Meteorological Conditions (in weather, or"IMC" and flying on instruments) using Station Keeping Equipment (SKE). SKEdepicts the other airplanes within a formation, and tells the pilot continuously where tofly to maintain the exact desired position, both vertically and horizontally, within theformation. The Air Force flies this mission at night as well, at a minimum altitude of 500feet above ground level (AGL), using visual procedures and night vision goggles("NVGs" light up the portion of the ground or sky the pilot is looking at by useof light amplification).

1.2.3 Rescue: In addition to flying typical combat aerial delivery missions,rescue units fly three other missions. The first is aerial refueling helicopters. Theserefuelings are conducted at altitudes as low as 1,000 feet and are done both day andnight, with night missions using NVGs. The second mission involves the deployment of liferafts and other materials to survivors at sea, by conducting air drop operations fromaltitudes of 150 feet above the water during the day and 500 feet above the water atnight. A third mission is search. Rescue crews routinely practice overwater and overlandsearches from altitudes as low as 500 feet. Over the years, rescue units have developed acapability to insert rescue forces long range into hostile territory. Their range andrefueling capabilities enable them to ferry rescue helicopters great distances forrecovery of these inserted rescue forces

1.2.4 Special Operations: Crews flying Special Operations aircraft fall intoseveral categories. Some fly aircraft in a rescue role similar to the rescue missiondiscussed above. Another Special Operations mission is to fly combat aerial delivery-likeoperations but in a more demanding environment. These aircraft are equipped with moreprecise navigation equipment, allowing lower altitudes during night operations. They arealso capable of landings on unlit landing zones and aerial refueling in flight (as thereceiver aircraft, thus extending their range).

1.2.4.1 When operating at night, gunships use sensitive optical and electronic sensorsto detect ground activities and direct a wide array of weapons to attack those targets.

1.2.4.2 "Commando Solo II" is the name for the psychological warfare missionflown by one Air National Guard unit. This mission involves orbiting near, or in somecases over, enemy territory to broadcast information or jam enemy operations. They operateat extremely high operational weights and can be refueled while airborne. These aircrafthave the highest empty gross weights of the fleet, owing to the broadcast equipment theycarry. When combined with their relative age, the requirement to refuel to near emergencygross weight limits for deployments on operational missions, and the high potential forRadio Frequency Interference (RFI) induced electrical problems, these factors identifythis mission as one associated with marked higher risk than others.

1.2.4.3. "Senior Scout" is another variant that serves as anintelligence-collection platform. The mission is accomplished by both active and ANGunits.

1.2.5 Polar Operations: One ANG unit flies C-130s equipped with skis for landingon snow and ice. This unit is principally responsible for support of Arctic and Antarcticoperations. These operations expose both aircraft and crews to the environmentalchallenges of extreme cold, variable weather and substandard landing zones.

1.2.6 Compass Call/Airborne Command, Control and Communications (ABCCC): Thesetwo missions have different operational roles but operate in basically the sameenvironment. The aircraft serve as a platform for communications activities. Operating ina "stand off mode" that generally places them adjacent to the battle area butnot directly in or over it, their operational environment is relatively benign. Themission’s chief drawbacks are its requirement to operate in proximity to unfriendlynations, its high aircraft operational weights, and the need to remain on station forextended periods of time.

1.2.7 Weather: The weather aircraft and crews, assigned to Air Force ReserveCommand (AFRC), are used for storm tracking and evaluation. By flying into hurricanes andtaking atmospheric readings at various locations within the storm, they obtain datacritical to improved weather forecasting. The principal dangers associated with thismission are weather related turbulence and lightning.

1.2.8 Other missions: Several units maintain a limited capability to conductunique operations.

1.2.8.1 Aerial Spray: These AFRC crews conduct low-level operations in rural and,occasionally, urban areas to dispense pesticides, oil dispersing agents and defoliants.These operations are conducted at altitudes as low as 100 feet and speeds of 125 knots.This operation is only conducted in daylight and in good weather.

1.2.8.2 Modular Aerial Firefighting System: This mission, flown by both Air NationalGuard and Air Force Reserve Command crews, involves dropping fire retardant foam on forestfires. This operation is conducted at altitudes of 150 feet, 130 knots. The greatestdemand on the crews and aircraft is operation in heavy smoke that reduces visibility inrugged terrain.

1.2.8.3 Space Shuttle Support: One rescue unit has the principal responsibility tosupport rescue efforts for every NASA Space Shuttle launch. The HC-130 serves as acommand-and-control platform, a jump platform, and an air refueling platform for rescuehelicopters.

Section 2.0

Aircraft Systems

2.1 Introduction

2.1.1 The team reviewed all major C-130 aircraft systems with specific emphasis onsafety-related issues or deficiencies that exist. This section includes a description ofeach system, safety issues that surfaced during the review, and the corrective actionstaken or that needed to be taken to mitigate the issues or deficiencies noted.

2.1.2 Special emphasis was placed on possible causes of uncommanded power reductions.The team accelerated the Failure Modes, Effects and Criticality Analysis (FMECA) of thesynchrophaser and required specific ground and flight tests to be conducted.

2.1.3 Prior to the establishment of the BAR, C-130 systems were in the process of beinganalyzed through the FMECA process to uncover hypothetical failure modes. It analyzesdesign and performance data to determine how the targeted systems perform their intendedfunctions, as well as whether those systems have unrecognized effects or synergisticinteractions with related systems. It then extrapolates the relative severity,probability, and worse-case impact of each identified failure mode. More simply put, eachpiece or part of the aircraft’s systems are being evaluated to determine what itsfunction is, how many different failure modes each piece or part can have, and how eachfailure mode effects both the systems and the aircraft as a whole.

2.1.4 The aircraft fuel system was subjected to ground and flight tests at Edwards AFB.The test objectives used were designed to see if the system could theoretically functionin ways not previously recognized. These objectives took into account both systems andaircraft design, as well as human factors. The team felt the dual approach of systemsanalysis and aircraft ground and flight tests was the best way to evaluate which system(or combination of systems) malfunctioning can result in a power-loss.

2.2 Structures

2.2.1 The airframe subsystem is the "skeleton" of the aircraft and supportsvarious flight and landing "loads" (i.e., the term used for stresses put on theairplane on the ground or in flight). The airframe subsystem is comprised of four majorstructural elements: the wing, fuselage (i.e., the body of the aircraft which actuallycarries the passengers and cargo), empennage (the "tail section" of theaircraft), and the landing gear. The primary purpose of the wing is to generate thelifting force needed to hold up the fuselage in flight. The fuselage structure must alsosupport cargo and pressurization load stresses, as well as the load stresses beingtransmitted from the wings and from the empennage. The empennage structure transmits andcarries the same type of load stresses as the wings, except that they are smaller andserve to keep the airplane stable around the vertical and lateral axes in flight. Last,the landing gear absorb the shock and vibration load stresses that occur as a result oftaxiing, takeoff, and landing. During the BAR’s review of C-130 flight safety, theynoted no flight safety concerns related to the aircraft’s structure that were notbeing addressed by the C-130 SPO.

2.3 Propulsion (Propellers & Engines)

2.3.1 The C-130 is powered by either four T56-A-7B or T56-A-15 engines. The majorcomponents of the engine are the power section, extension shaft assembly, and thereduction gear assembly.

2.3.2 Power Section. The power section of the engine has a single entry, 14-stageaxial-flow compressor, a set of 6 combustion chambers, and a 4-stage turbine. Mounted onthe power section are an accessories drive assembly and components of the engine fuel,ignition, and control systems (the engine fuel system is described in detail later in thissection). The ignition system is a high-voltage, condenser discharge type, consisting ofan exciter, two igniters, and control components. The ignition system is powered by theessential DC bus. The system is controlled by the speed-sensitive control through theignition relay, which turns it on anytime the engine RPM is between 16 and 65 percent. Amanifold bleeds air from the compressor for airplane pneumatic systems. Anti-icing systemsprevent accumulation of ice in the engine inlet air duct and the oil cooler scoop. Fuelflows into the combustion chambers and is burned, increasing the temperature and energy ofthe gases. The gases pass through the turbine, causing it to rotate and drive thecompressor, propeller, and accessories.

2.3.3 Extension Shaft Assembly. The extension shaft assembly consists of two concentricshafts and the torquemeter components. The inner shaft transmits power from the powersection to the reduction gear assembly. The outer shaft serves as a reference shaft forthe torque indicating system.

2.3.4 Reduction Gear Assembly. The reduction gear assembly reduces the high speed ofthe engine (13,820 RPM) to the lower speed needed by the propeller (1,020 RPM). Thereduction gear contains a reduction gear train, a propeller brake, an engine negativetorque control system, and a safety coupling. The reduction gear train is in two stages,providing an overall reduction of 13.54 to 1 between engine speed and propeller speed.

2.3.5 Related Deficiencies and Concerns:

DEFICIENCY: In the event of total AC electrical failure or flameout of all fourengines in-flight, the engine ignition system cannot be powered from the aircraft battery.

ON-GOING RESOLUTION: The BAR supports development and implementation of a C-130 modification which would allow the ignition system (essential DC bus) to be powered from the aircraft battery in-flight in the event of total AC electrical failure or flameout of all four engines.

2.3.6 Each engine is equipped with a Hamilton Standard four-blade, electro-hydromatic,full feathering, reversible-pitch propeller. The propeller operates as acontrollable-pitch propeller for throttle settings below FLIGHT IDLE and as aconstant-speed propeller for throttle settings of FLIGHT IDLE or above. The majorcomponents of the propeller system are the propeller assembly, the control system, thesynchrophasing system, and the anti-icing and de-icing systems.

2.3.6.1 Propeller Assembly. The propeller assembly consists of the actual propellerblades, the barrel assembly (which retains the blades and also contains the pitch lockassembly), and the dome assembly (which contains the pitch changing mechanism and the lowpitch stop assembly).

2.3.6.2 Control Assembly. The control assembly is mounted just behind the propellerassembly but does not rotate. It contains the oil reservoir, pumps, valves, and controlcomponents which supply the pitch changing mechanism with hydraulic pressure to change thepropeller blade angle. All mechanical and electrical connections necessary for propelleroperation are made through the control assembly. The mechanical connections are for theengine control system and the negative torque signal (NTS) system. The electricalconnections are for oil level indications, pulse generator coil, auxiliary pump motor,synchrophasing system, NTS and feather switches, anti-icing and deicing systems and theelectrical feathering system. The valve housing is the "brain" of the propellerand contains the fly weight speed sensing pilot valve, feather valve, feather solenoidvalve, and feather actuating valve. The speed of the propeller is controlled by the flyweight speed sensing pilot valve. The valve is controlled by the mechanical action of theflyweights opposing the force of the speeder spring. Under normal conditions thepropellers are rotating at 100% of its design speed (1,020 RPM) or "on-speed".When the propeller is in an on-speed condition, the metered hydraulic pressure equals thatrequired to maintain the required blade angle. When an overspeed condition occurs, the flyweight force overcomes the speeder spring force, and the pilot valve moves to porthydraulic pressure to increase the blade angle which causes the propeller to slow down. Ifthe propeller slows down below the governed speed, the force of the speeder springovercomes the force exerted by the fly weights, and the pilot valve moves to porthydraulic pressure to decrease the blade angle, which allows the propeller to increasespeed. The action of the fly weight speed sensing pilot valve is the primary means ofcontrolling the RPM of the propeller and is always attempting to maintain 100% (1,020 RPM)in flight.

2.3.6.3 Synchrophasing System. The propeller mechanical governor will hold a constantspeed in the flight range, but throttle changes will cause the governor to overspeed orunderspeed slightly while trying to compensate for the change in power. The synchrophasingsystem assists the mechanical action of the fly weight speed sensing pilot valve. Thesynchrophaser provides speed stabilization, throttle anticipation, and synchrophasing. Thespeed stabilization circuit stabilizes the mechanical governor when the propeller governorcontrol switch is in the NORMAL position by sending a signal to the speed bias servo motorto change the speeder spring compression. Throttle anticipation stabilizes the propellerspeed during rapid movement to the throttle when the propeller governor control switch isin the NORMAL position. Rapid throttle movement sends an amplified signal to the speedbias servo motor to change speeder spring compression. The synchrophasing system acts tokeep all the propellers turning at the same speed, and it maintains a constant rotationposition relationship between the blades to decrease vibration and to lower the noiselevel. The system uses either the number 2 or the number 3 engine as the master engine,and relates the blade position of the other three propellers to the master. The bladeposition of a slave propeller is changed by moving the pilot valve to increase or decreasethe speed of the engine. The synchrophasing circuit determines blade position by comparingan electrical pulse generated by each slave propeller to a pulse from the masterpropeller. In normal governing and synchrophaser modes, the synchrophaser can only changethe RPM of the propeller approximately 2.5%. Mechanical stops in the propeller valvehousing prevent the RPM from decreasing more than 4% (below 96%) or increasing more than6% (above 106%).

2.3.6.4 Anti-Icing and Deicing System. The propeller anti-icing and deicing system ismade up of resistance-type heating elements which are incorporated on the leading edge andfairing of each blade and the entire spinner assembly for anti-icing. Continuousanti-icing heaters cover the front portion of the spinner assembly and the entireafterbody assembly. Cyclic deicing heaters cover the remainder of the spinner frontsection, the spinner rear rotating section, the spinner plateaus, and the blade leadingedge and fairing. Power from the aircraft electrical system is transmitted through a brushhousing assembly through rotating sliprings to the anti-icing and deicing elements.

2.3.7 Rollback. The term "rollback" has been used for several years todescribe an event in which multiple engines experience a sudden, relatively smallreduction in engine speed, uncommanded by the crew and with no prior indications of engineproblems such as fluctuating fuel flow or turbine inlet temperature (TIT). Rollbacks havehistorically been associated with the synchrophaser or electrical system problems, such aslow voltage or electromagnetic interference (EMI), which can affect synchrophaseroperation.

2.3.7.1 During a rollback, affected engines respond essentially simultaneously. Somerollbacks are momentary, (i.e. the RPMs pull back for a few seconds) and then recoverwithout crew intervention. Some rollbacks persist and do not recover until the correctiveprocedures are performed. However, in both instances, the engines exhibit relativelystable operation except for momentary torque and RPM changes. In other words, the torqueand RPM changes are not accompanied by significant variations in fuel flow and TIT.

2.3.7.2 Since the synchrophaser has no direct control over the power output of theengine and has such limited RPM control authority, it cannot by itself cause large,unstable, erratic variations in fuel flow, TIT, torque, etc. These are indications thatsomething else is affecting the power output of the engine, such as a fuel system problem.However, if the engine selected as "master" (either number 2 or number 3 engine)is experiencing problems for any reason which reduces its power enough to affect its RPM,the other three "slave" engines will be driven by the synchrophaser into smallvariations, as it tries to keep them in phase with the malfunctioning engine. If thisshould happen, the RPMs of the three remaining engines will only be driven down by amaximum of 2.5%.

2.3.7.3 Ground tests during earlier synchrophaser investigations confirmed thaterroneous output signals to the propeller governors had little to no effect on actualengine power output. One possibility tested was that those electrical disturbances knownto produce erroneous synchrophaser signals could also independently affect the temperaturedatum (TD) amplifier, causing uncommanded power reduction unrelated to synchrophaser RPMreduction. The laboratory instruments revealed these electrical disturbances did noteffect actual engine power output, but did produce erroneous aircraft instrument readings.

2.3.7.4 Crew reactions to rollbacks and power fluctuations have varied considerably.Some report that the event is no more than a mild annoyance; others say it really getstheir attention. However, the measurable effect of a rollback on the flying qualities ofthe aircraft has proven to be very small. Rollbacks have been extensively investigated,both by gathering data on actual incidents and by duplicating the events during enginetest cell runs and flight tests. As a result, rollback causes have been identified andcorrective actions implemented which have dramatically reduced both the frequency and themagnitude of the events.

2.3.7.5 Engine rollbacks can be caused by internal synchrophaser failure, resulting inerroneous output signals. Since the synchrophaser was redesigned (from vacuum tube tosolid state technology), this is now a rare event. The new synchrophasers containsafeguards designed to limit the magnitude of erroneous signals if the unit did fail.Although reliability increased with the solid state units, they did exhibit vulnerabilityto electrical power disturbances and EMI, or "noise," from other aircraftelectronic systems, notably the high frequency (HF) radio. Several equipment modificationshave been made which have been effective in reducing the frequency and impact of theseevents. Some wires have been shielded, components redesigned and additional componentsinstalled to stabilize the synchrophaser system. As shown in the failure history, (seeFigure 5-3), these changes have been effective in reducing rollback occurrence. Acontinuing problem is the susceptibility of the aging synchrophaser wiring bundles to EMI.

2.4 Fuel System

2.4.1 Aircraft Fuel System. The primary purpose of the fuel system is toefficiently distribute fuel to its engines. There are four main fuel tanks and twoauxiliary tanks located in the wings. The main tanks are numbered 1 through 4, from leftto right. The auxiliary tanks are located in the center wing. Two external tanks areinstalled on pylons under the wings (C-130E/H aircraft and their variants). Tanker/rescueand other special mission aircraft are also equipped with one or two internal fuselagetanks. The fuel tanks are vented and fuel is displaced by ambient air as the tanks empty(fuel is displaced by cabin air for the fuselage tanks). Fuel management is controlled bythe flight engineer through the overhead fuel control panel located on the flight deck(see Figure 2-1). C-130 tanker aircraft have an additional air refueling panel whichcontains the controls and gauges for the fuselage tanks that is also controlled by theflight engineer (see Figure 2-2). By positioning the switches on the panel(s), the flightengineer can control how fuel is used during flight as well as manage refueling anddumping operations.

C-130 Broad Area Review (1)

Figure 2-1: Typical Overhead Fuel Panel, located on thecockpit ceiling between the pilots

2.4.1.1 All fuel tanks are interconnected by two pipes or "manifolds". One iscalled the crossfeed manifold, the other is the refuel/dump manifold. These two manifoldscan also be interconnected. By selectively opening or closing various valves andactivating or deactivating pumps, fuel can be pumped from any tank to any engine theflight engineer requires.

2.4.1.2 When each engine is fed by its corresponding fuel tank (number 1 tank suppliesfuel to the number 1 engine, etc.), it is described as "tank-to-engine"configuration. This is the normal fuel system configuration used for take-off and landing(see Figure 2-3). The "tank-to-engine" configuration requires the partitioningof the crossfeed manifold through the use of shut-off valves and the use of eachtank’s boost pump to provide positive fuel pressure.

C-130 Broad Area Review (2)

Figure 2-2: Typical HC-130N/P Auxiliary Fuel Panel

C-130 Broad Area Review (3)

Figure 2-3: Typical HC-130N/P Fuel System inTank-to-Engine Configuration

2.4.1.3 When the crossfeed manifold shutoff valves are opened and fuel is drawn fromthe external, fuselage, auxiliary, or other main fuel tanks, it is described as crossfeedoperation (see Figure 2-4). This is usually used during cruise flight to utilize the fuelcontained in the auxiliary, external, and fuselage tanks.

C-130 Broad Area Review (4)

Figure 2-4: Typical HC-130N/P Fuel System in CrossfeedConfiguration

2.4.1.4 The boost pumps for the external, fuselage, and auxiliary tanks have a higheroutput pressure (28-40 psi.) than the main tank boost pumps (15-24 psi.). This designfeature allows the main tank pumps to remain on, so that in the event that fuel is notdelivered from any of these tanks, the main tank boost pumps deliver the required fuelimmediately. In each case, the higher pressure over-rides the weaker main tank boostpumps. When the selected tank is empty, the main tank pumps assume the task of supplyingfuel to the crossfeed manifold. In an emergency, fuel can be dumped overboard by usingdump pumps located in each tank and directed through the dump manifold to the dump mastslocated in each wing tip.

2.4.1.5 A safety improvement implemented by the Air Force has been the transitioningfrom JP-4 fuel to the less volatile JP-8. Some slight problems have surfaced during thisfuel transition, such as small fuel leaks which have been attributed to o-ring shrinkage.The solution has been to replace o-rings with new ones on an "as required"basis. Additionally, improvements in tank sealing technology have drastically reduced thenumber of external fuel leaks attributed to improper fastener installations or jointsealing interfaces.

2.4.1.6 Over the years, the basic fuel system has been modified and/or enhanced,depending on the mission design series (MDS). Some C-130s have fuselage tanks to extendtheir range and/or allow other aircraft to be refueled in flight. Additionally, someaircraft have been modified to receive fuel in flight from an aerial tanker.

Fuselage Tanks. The fuselage tanks were added to the aircraft to enhance thefuel capacity/range. Consisting of one or two 1800 gallon cylindrical tank/s they aremounted in the cabin section of the fuselage, near the aircraft’s center of gravity.The fuselage fuel tank has either a single or dual pump configuration. The pumps are ratedfor 28-40 psi. The single and dual pump tank configuration is not consistent from aircraftto aircraft, since these tanks are removed and replaced for mission purposes andmaintenance requirements routinely. Depending upon mission requirements, some planes willfly with zero, one, or two fuselage tanks. The fuselage tanks are plumbed into theaircraft’s refuel/dump manifold. The right external dump valve is controlled with theright external crossfeed valve switch, connecting the dump manifold to the crossfeedmanifold. This allows fuel from the fuselage tanks to be routed to the crossfeed manifoldfor engine consumption. The fuselage fuel tanks vent system is unique to the otheraircraft fuel tank vent systems. Because the fuselage tank resides in the aircraft’spressurized cabin section, the fuselage tanks internal space is maintained at cabinpressure to prevent structural damage to the tanks. This is accomplished by a vent systemthat allows cabin air to enter the tank when the aircraft is pressurized or fuel is pumpedfrom the tank. The vent system also allows pressure in the tank to be vented outside theaircraft when the aircraft is depressurized or fuel is pumped into the fuselage tank.

2.4.1.7 The C-130 basic fuel system has required no major modifications to resolvesafety concerns. However, slight modifications such as the addition of the 1360 gallonexternal pylon tanks (installed on C-130E/H aircraft and their variants), valverelocations and operations, and plumbing routing, have been incorporated to extend thecapabilities of the C-130. The C-130 fuel system experiences normal wear and tear whicheventually requires individual components to be repaired/overhauled and/or replaced. Forexample, boost and dump pumps fail; o-rings and couplings leak; gate, and butterfly valvesfail; and fuel system plumbing may become damaged. Occasionally, isolated fuel systemrelated problems have required depot engineering assistance.

2.4.1.8 Failure of the fuel system to provide fuel to the engines centers around twoconditions:

Insufficient fuel getting to the engine burner cans - Insufficient fuel could bethe result of a fuel leak, inadvertent fuel dumping, fuel system or engine componentfailure, or allowing the tank to be emptied without other tank boost pumps on or withoutswitching to a fueled tank. Failure to prime the fuel manifolds can result in existing airin the fuel manifolds being sent to the engines. The consequence of these conditions maybe an engine flameout.

Contamination - Contamination interferes with the engine fuel system’sability to deliver a sufficient quantity of fuel to the engine combustion chambers whichcan reduce the power output of the engine. The circumstances leading to this arecontaminated fuel introduced at the last refueling, fuel becoming contaminated due toin-tank debris from maintenance or internal deterioration of the tank or fire suppressionfoam, or water due to condensation or rain entering through filler caps. In extreme cases,such contamination may clog filters and their associated bypass valves, resulting inengine flameout.

2.4.2 Engine Fuel System

C-130 Broad Area Review (5)

Figure 2-5: C-130 Engine Fuel System

2.4.2.1 The main components of the engine fuel system consists of the fuel pump, fuelcontrol, temperature datum (TD) valve, and fuel nozzles, along with drain valves and twofuel filters. Fuel flow through these components is shown in Figure 2-5.

2.4.2.2 Fuel supplied from the aircraft fuel tanks is delivered to an engine-driven,two-stage, gear type fuel pump. In the event of the failure of one stage of the pump, thetwo-stage design ensures the engine will be supplied with sufficient fuel.

2.4.2.3 The fuel control is a hydro-mechanical metering device designed to supply acontrolled fuel flow to the engine during all operating conditions. Located on the engine,the fuel control measures the RPM of the engine, the inlet air temperature, inlet airpressure, and throttle position. Fuel metered by the control is equal to enginerequirements, plus an additional 20%, which is for the use of the TD valve, a part of theTD system. There are six fuel nozzles mounted in the "diffuser case" of eachengine. One fuel nozzle extends into the forward end of each of the six combustion liners.Looking like large metal cans with holes punched regularly around their sides to carefullycontain the flames, two of these opposing liners have igniter plugs to ignite the fuelduring engine start. As the fuel nozzles disperse the fuel in a fine spray to maximizecombustion efficiency within the engine, interconnecting tubes between the liners spreadthe flame and assure complete combustion within.

2.4.2.4 The TD valve is an electrically operated fuel-trimming device. All fuel flowingfrom the fuel control must pass through it before being sent to the fuel nozzles. Sincethe TD valve receives 120% of engine fuel flow requirement from the fuel control, somefuel must be bypassed by the valve. Fuel in excess of that required by the engine isbypassed and returned to the inlet of the fuel pump. When only that 20% surplus isbypassed, this is known as a "NULL" condition. When less than 20% is beingbypassed to the fuel pump, this is known as a "PUT" condition. When more than20% is being bypassed, this is known as a "TAKE" condition.

2.4.2.5 A small drive motor operates a piston inside the TD valve which adjusts howmuch fuel is ultimately passed on to the fuel nozzles. This drive motor is controlled by asignal from the TD amplifier (TD amp).

2.4.2.6 The TD system works to help keep the engine running at the temperature andpower setting the pilot selects when he moves the throttles. It also allows the engines toburn several different kinds of fuel. The TD amp receives a temperature signal from 18engine "thermocouples." These are metal probes which accurately measure hightemperatures within the engine. They are mounted at the inlet of the turbine section ofeach engine, just behind where the fuel is burned and the resultant gases are near theirhottest point. The TD amp compares an average of these 18 signals from the inlet of theturbine with the "reference signal" in the TD amp (for start temperaturelimiting protection), or from the temperature setting commanded by the pilot’sthrottle movements. The throttle’s signal is generated through a"potentiometer" (i.e., a rheostat) in the "throttle coordinator,"which is also on the side of the engine. The signal from the potentiometer corresponds tothe position of the engine throttle lever. When the temperature signal from thethermocouples matches the reference signal, the TD amp sends no signal to the TD valve,and the valve remains in the "NULL" Position. If the temperature signal isgreater than the reference signal, the TD amp sends a signal to the TD valve to"TAKE" fuel. If the temperature signal is less than the reference signal, the TDamp sends a "PUT" signal to the TD valve. In this fashion, through thousands ofsmall corrections, the TD system constantly works to keep the engine operating at thedesired temperature.

2.4.2.7 Related Deficiencies and Concerns

DEFICIENCY: Some fuselage tanks have only one fuel pump.

ON-GOING RESOLUTION: Continue installing pumps until all fuselage tanks have two pumps.

DEFICIENCY: In the event of a main tank failure, T.O. 1C-130(H)H-1, page 3-23allows the use of the dump pump from the same main tank to crossfeed to its respectiveengine. This is an adequate procedure but its use should be discontinued before the dumppump inlet is uncovered. According to T.O. 1C-130(H)-1, page 1-47, this occurs atapproximately 1,500 to 2,100 lbs, depending upon the specific type of C-130.

RECOMMENDATION: The BAR recommends that Air Force establish a fuel quantity limit for using the above procedure and revision of all affected C-130 T.O.s accordingly.

2.5 Electrical System

2.5.1 Four engine-driven alternating current (AC) generators and an auxiliary generatorpower the AC electrical system of the C-130. On aircraft prior to tail number 74-1658, anauxiliary AC generator is driven by an air turbine motor (ATM) which is operated byhigh-pressure air from either the gas turbine compressor (GTC), or from an operatingengine. The GTC cannot be operated in-flight, but the ATM can be operated to supply ACpower if the bleed air manifold is pressurized. Newer aircraft received an auxiliary powerunit (APU) which can be operated in flight. The APU generator is directly splined to theAPU and does not rely on bleed air to operate.

C-130 Broad Area Review (6)

Figure 2-6: C-130 AC Bus Power Sources

2.5.2 The AC generators are connected through transfer contactors (relays) to four ACbuses: the left hand AC bus, the essential AC bus, the main AC bus, and the right hand ACbus. The transfer system is automatic and operates in such a manner that any combinationof two or more engine-driven AC generators will power all four AC buses. With only one ACgenerator supplying power, only the essential and main AC buses will be powered. In theevent of complete loss of all AC generators, none of the AC buses will be powered.Operation of the APU or ATM generator will supply power only to the essential AC bus.Combinations of operating generators and the buses they power are shown on the AC buspower sources chart in Figure 2-6 above.

2.5.3 The normal source of direct current (DC) power is four transformer-rectifier (TR)units. These units change the three-phase AC power from the AC generators to 28 volt DCpower. A 24-volt battery is provided as an emergency source of DC power. Two of the TRunits are connected to the essential AC bus with their output supplying power to theessential DC bus. The other two TR units are connected to the main AC bus with theiroutput supplying the main DC bus. In the event of total AC power failure, the essential DCand main DC buses will also lose power. In this case the only remaining power source isthe aircraft battery which will power the battery bus and the isolated DC bus to providebasic instruments and communication for the flight crew.

2.5.4 Related Deficiencies and Concerns

DEFICIENCY: Configuration Control - Over the previous 10 years, detailed controlover and knowledge of the exact configuration of each aircraft has been lost. This is theresult of having many different C-130 users, several diverse missions, and no cohesiveprogram to force compliance upon the various operators. As a result, these aircraft weremodified in a less than stringently controlled environment, and by modification teamswhich did not always precisely follow the modification drawings. Changed under provisionswhich allowed the C-130 manufacturer to deliver new aircraft with prior modificationsinstalled, their users sometimes preferred cockpit equipment arrangements different fromthe standard C-130 configuration, but did not fully document these changes. In addition,there are users who modified the aircraft to meet mission needs without documenting thechanges, as well as test agencies who did not document their modifications.

ON-GOING RESOLUTION: The BAR supports the C-130 Systems Program Office efforts to institute a configuration management program for USAF C-130 users. Through various directives, the users are required to document all changes and process those changes through a configuration-control process that includes air-worthiness and major command concurrence. Additionally, a digital photographic record and digital location matrix for each aircraft are being gathered. These data are the core of the continued documentation of aircraft configuration as aircraft modernization progresses. Currently, C-130s are undergoing modification to upgrade the electrical system to provide clean avionics power, install defensive avionics, replace unsupportable auto-pilots and install a global positioning system (GPS). These modifications are prerequisites for future upgrade efforts, which will include the recorders, the second INU, the radar, a modernized avionics suite and flight station. Future plans include replacement of the flight data and cockpit voice recorders, replacement of the radar, and installation of a second inertial navigation unit (INU) to replace vertical gyros and compasses.

DEFICIENCY: Many aircraft have wiring that is reaching the end of its servicelife. Additionally, avionics wiring has been damaged during numerous modifications whichinstall or relocate systems. This wiring degradation increases its susceptibility to EMI.Also, the structure of these aging aircraft has lost some electrical bonding propertieswhich are essential to avionics operation. The C-130 uses the aircraft structure as groundreturn for the electrical equipment. Installation of the structural components must beaccomplished to ensure that each structural piece is electrically the same (zero volts).Electrical systems behave erratically when the ground potential is not zero volts.

RECOMMENDATION: The modifications mentioned above serve to replace most of the electronic wiring and allow the remaining wires to be inspected and repaired or replaced. The immediate need for correcting wiring short-comings, relative to some rollback scenarios, is for the replacement of synchrophaser interface wiring bundles. Also, the aircraft can be improved by rewiring wings and wheel wells. Many maintenance procedures address bonding/grounding integrity; however, both will be assessed and improved during the modernization effort. A continuing problem is the susceptibility of the aging synchrophaser interface wiring bundles to EMI. The BAR recommends the C-130 System Program Office propose a modification to replace the synchrophaser interface wiring bundles on all C-130 aircraft.

DEFICIENCY: Digital Flight Data Recorder (DFDR) - The current DFDR is anunreliable circa-1970s magnetic tape system which has limited channels for recording data.Data recording captures only 25 hours of data. Verification that the system is functionalis an arduous two-month process involving shipment of verification tapes to a remotereading site. This system and the accompanying cockpit voice recorder (CVR) areinconsistent with the needs of the safety boards and are incompatible with the projectedconfiguration of the modernized instrument system.

ON-GOING RESOLUTION: The BAR supports the C-130 SPO efforts to modernize the DFDR/CVR in two phases. First, the current DFDR will be replaced with a form, fit, and function solid state recorder which will have additional parameters added. This system will use a solid state recording media which can be read by the using unit to verify proper function. Second, as a part of the modernization of the aircraft, a large parameter capacity system will be installed to accomplish data and voice recording. This system will be compatible with data buses to allow parameters to be recorded directly.

DEFICIENCY: The loss of the DFDR and Cockpit Voice Recorder (CVR) when the poweris lost from the last engine generator is considered a deficiency because there is noavailable record of crew actions or aircraft performance from that point until ground orwater impact.

RECOMMENDATION: This report supports C-130 SPO efforts to develop and implement a modification which would provide power to the DFDR and CVR during battery only operation.

DEFICIENCY: The exact desired parameters for DFDR recording need to be definedso as to ensure the DFDR records the essential performance data necessary for post-mishapanalysis, considering FAA and NTSB guidelines and/or recommendations.

ON-GOING RESOLUTION: This report supports Air Force Safety Center efforts to provide a list of these parameters.

DEFICIENCY: Review of the aircraft configuration revealed that the ESUmodification has configured the generator controls such that, on some E-model aircraft,the crew must shut down the engine when its generator has failed.

RECOMMENDATION: The BAR recommends the C-130 SPO develop and implement a modification which would install generator disconnects or bearing failure lights in all ESU aircraft.

DEFICIENCY: The crew cannot operate the fuel valves, nor can the engine ignitersbe powered when the aircraft is in an airborne, battery-only condition.

ON-GOING RESOLUTION: The BAR supports development and implementation of a C-130 modification which would bypass the touch-down relay to allow the DC isolated bus to power the DC essential bus, thus allowing fuel valves and igniters to function. This effort would be accomplished concurrently with the existing modification to add an additional reverse current relay. This will provide the ignition source to restart the engines if fuel is available. The BAR does not consider the APU replacement a safety issue since the above provides an alternative way to provide restart capability to an aircraft with windmilling engines. This proposal is currently being evaluated by WR-ALC and is awaiting funding. If funded, it should be complete by Dec 31, 1998. The BAR reviewed the restart capability of the C-130. This is identified as a deficiency elsewhere in this report. Those C-130 aircraft manufactured before 1974 utilized a GTC and ATM to provide limited aircraft electrical power on the ground and in certain in-flight emergencies. The ATM uses bleed air from either the GTC (ground use only) or engines to produce electricity. In the air, the GTC can not be operated so only bleed air from engines can be used to power the ATM. This design feature means that with no engines operating, the ATM can not provide any electrical power. The post-74 aircraft had an APU added to provide electrical power in the same situations as the ATM. It has the additional capability of utilizing external air flow and can provide electrical power with no engines running. The Air Force has considered a retrofit to replace the GTC/ATMs on pre-74 aircraft, with an APU. The decision process involved a risk assessment and cost considerations. The high reliability of the C-130, the absence of any four engine flame-outs in over 24 million flight hours, and the high cost ($1,200,000 per aircraft) argued against the modification. The Air Force is currently proposing a broad ranging upgrade to the C-130 that includes the APU. This modification is driven by international airways conventions, required avionics upgrades, configuration control, and maintainability issues. Including the APU in the modification program substantially reduces the cost of doing the APU upgrade independently.

2.6 Pneumatic System

2.6.1 The pneumatic system provides compressed hot air "bled" from the enginecompressors to operate a number of aircraft systems, such as engine ground starting,anti-icing, air conditioning and heating, and pressurization. This air is distributed tovarious locations within the aircraft via metallic ducts. Over time, these ducts haveexhibited a susceptibility to corrosion, resulting in a rupture of the ducting. Because ofthe heat associated with a bleed air leak, significant damage to the aircraft and othersystems can result from a duct failure. As a result of several previously documented ductfailures, a program is currently in progress to replace these bleed air ducts with onesmade from a more corrosion-resistant nickel alloy. At this time, all of the flight-safetycritical ducts have been replaced in all C-130 aircraft, and many other less criticalducts have been or are scheduled to be replaced by the C-130 SPO.

2.6.2 During the review of C-130 flight safety, the BAR noted no flight safety concernsrelated to the aircraft’s pneumatic system that had not been addressed.

2.6.3 Related Deficiencies and Concerns

DEFICIENCY: None.

2.7 Hydraulics System/Flight Controls

2.7.1 The C-130 has three separate 3,000 psi hydraulic systems: the booster, utilityand auxiliary systems. These systems are used to operate flight controls, cargo ramp anddoor, flaps, brakes, nosewheel steering, and the landing gear. The booster and utilitysystems are supplied by four engine-driven pumps mounted on each engine’s reductiongearbox. The auxiliary hydraulic system is supplied by an electrically driven pump. Forreasons of safety and system reliability, the flight control boost packs for the elevator,rudder, and ailerons are usually pressurized by both the booster and utility systems.Manual operation of the flight controls without any hydraulic assistance is also possibleduring an emergency.

2.7.2 During the review of C-130 flight safety, the BAR noted no flight safety concernsrelated to the aircraft’s hydraulic system that had not been addressed by the C-130SPO.

2.7.3 Related Deficiencies and Concerns

DEFICIENCY: None.

Section 3.0

Operations and Training

3.1 C-130 Technical Manuals ("Technical Orders")

3.1.1 Technical Orders apply to both operators (primarily flight manuals) andmaintenance personnel. The BAR will address the flight manual deficiencies in thissection. For the aircrew, the flight manual describes the aircraft and how to operate it.It is divided into several sections, including normal operations and emergency procedures.There is a second volume called the performance manual that contains detailed charts tocalculate aircraft performance for given conditions (e.g., takeoff , climb, cruise,descent, and landing). The aircrews need an accurate, easily read document for use inflight during normal, abnormal and emergency situations. The following problems wereidentified with the current manuals.

3.1.2 One flight manual had over 30 active operational supplements altering the basicdocument and requiring changes to be annotated by hand.

3.1.3 One operational supplement required over 12 hours to post, including write-inchanges to critical portions of the emergency section of the manual .

3.1.4 Some charts in the performance manual contain inaccurate data. For example,torque charts in the C-130H model performance manual are known to be in error by up to 7%.In other cases, aircraft configuration changes have outpaced performance manual updates,creating problems in many aircraft variants. One example of this resultant mismatch is thedrag index for the Commando Solo II aircraft. Finally, important charts are not available.For example, three- and four-engine climb gradient charts are needed for all models of theC-130.

3.1.5 Several variants of the aircraft have their own separate flight manual. This isnecessary because of unique aircraft differences. However, many identical systems in thesemultiple variants have different procedures mandated in their flight manuals. For example,some commands require engine run-ups while some don’t; some require top of theaircraft inspection and some don’t, some require a positive fuel flow check whilesome do not. There are individual units where flight manuals for the various aircraftdon’t match. In a special operations unit, with a mixture of specialized and basicC-130s, two separate commands manage their two flight manuals. Crews are expected to becompetent in both, but even the basic operating procedures are not the same.

3.1.6 All USAF C-130s do not use AF flight manuals. Some special mission aircraft useLockheed Technical Manuals (LTMs) for flight operations. Crewmembers assigned to theseunits also operate basic C-130s which use AF flight manuals. As in the above example,because the LTMs and the AF flight manuals do not always contain the same procedures orguidance, aircrew must use different procedures for identical systems depending upon whichtype aircraft they are flying.

3.1.7 These major operating procedure differences are a by-product of several commandshaving responsibilities for the flight manual content for the variants under theircontrol. A single lead agency, responsible for conformity, would reduce this problemsignificantly.

3.1.8 Related Deficiencies and Concerns

DEFICIENCY: Technical Orders currently contain too many supplements that are toolarge and laborious to incorporate.

RECOMMENDATION: The BAR recommends that the Air Force update, consolidate and standardize technical orders, flight manuals, and published guidance, and to limit the number of write-in changes that can be introduced before a manual must be replaced.

DEFICIENCY: Non-standardized procedures among the many versions of the flightmanuals.

ON-GOING RESOLUTION: The C-130 Program Office will host a meeting, tentatively scheduled for 2-13 February 1998, with all C-130 major commands to identify the non-standardized procedures and reach agreement on standard procedures.

DEFICIENCY: Performance manual charts do not match aircraft performance, norcontain needed information.

ON-GOING RESOLUTION: The C-130 Program Office plans to correct the performance manuals and will get this effort underway in the third quarter of fiscal year 1998.

3.2 Aircrew Life Support Equipment

3.2.1 The BAR reviewed the guidance directing the placement of life support equipmentcarried on C-130 aircraft. They found different commands had different requirements. Whenthe CONUS CAD C-130s moved from ACC to AMC there were significant guidance changes.Currently AFI 11-302 Vol. 1 is in coordination and will provide Air Force-wide guidancestandardizing required C-130 life support equipment. The BAR is comfortable that the draftAFI 11-302 addresses the significant issues and will provide adequate guidance andstandardization. It is due out in early 1998.

3.3 Ditching/Bailout

3.3.1 A review of ditching/bailout information and procedures in various flight manualsreveals a significant variation in implied survivability of a ditching maneuver. FlightManuals managed by AMC and Air Force Special Operations Command (AFSOC) have significantlydifferent guidance. Specifically, T.O. 1C-130H-1 managed by AMC states:

"…ditching of transport-type airplanes can usually be accomplished with a high degree of success." Pg. 3-71 under "ditching."

3.3.2 However T.O.-1C-130(A)U-1, managed by AFSOC, states

"…ditching of the AC-130U can be accomplished with a low probability of aft crew member survivability. The flight deck may be survivable, but ditching should be considered an absolute last resort for any crewmember." Pg. 3-90 under "ditching."

3.3.3 Another apparent contradiction appears in T.O. 1C-130H-1 pg. 3-71 under"ditching characteristics"

"…Reasonably high probability that the airplane can be landed on water without major collapse of structure or a sudden rush of water into occupied compartments."

3.3.4 Compare this citation to the following language found in T.O. 1C-130(A)U-1 pg.3-90 under "ditching characteristics"

"…reasonably high probability the aircraft structure will collapse followed by a sudden rush of water into occupied compartments."

3.3.5 These two flight manuals addressing the same subject portray vastly differentprojections of success for a ditching attempt. Of similar concern the diagram on page 3-76of T.O. 1C-130(H)H-1 titled "Emergency Exits-Water" depicts a fully intact C-130floating in the water. None of the aircraft in the three most recent C-130 ditchingssurvived intact. The referenced diagram gives support to the idea of a survivableditching.

3.4 C-130 Aircrew Training

3.4.1 Categories of Training C-130 Aircrew Training falls into four majorcategories: Initial C-130 Qualification Training, Initial C-130 Mission QualificationTraining, Continuation Flying Training, and Continuation Ground Training.

3.4.1.1 Initial C-130 Qualification Training. This is the first formalAir Force training course a prospective C-130 crew member will attend and is normallyconducted at Little Rock Air Force Base in Arkansas. It consists of classroom, aircraftsimulator, and flight training. The classroom phase covers the basics of aircraft systemsand checklist procedures. During the simulator phase, the new pilots and engineers learnto integrate their systems knowledge and checklist use into a coordinated crew effort.They practice both normal and emergency procedures, simulate flying entire missions, anddevelop an understanding of crew coordination, learning to work together effectively in acrew environment. The two combined phases consist of 210 hours in the classroom, and 36 hours in the simulator. Navigators receive training in theaircraft systems navigation simulator to become familiar with how to guide the aircraftusing its self contained navigation system (SCNS), and the other compass and navigationalaids on board. During the flight phase, the new crewmembers will practice basic aircraftmaneuvers, both on the ground (i.e., engine start, taxiing, and backing), and in the air(e.g., takeoffs and landings, en route navigation, instrument approaches, and visualtraffic patterns). Flight hours logged vary by position, with new pilots accumulating themost hours (approximately 60 hours per pilot) during these courses owing to the necessityto develop their flying skills. The successful completion of this training qualifies anindividual to fly "basic air-land missions," i.e., those requiring getting fromone point to another while carrying passengers and/or cargo, in the C-130. Little Rock AirForce Base historically produces approximately 2,400 graduates (cumulative, all crewpositions) per year.

3.4.1.2. Initial C-130 Mission Qualification Training. This training initiallyqualifies crewmembers in a specific operational mission of the C-130 (such as airdrop,short field landings, rescue, special operations, firefighting, electronic combat,psychological operations (PSYOP), and others. Like initial qualification, this phase alsoconsists of classroom, simulator, and flight training. For those trained in the combatdelivery mission, which comprises the largest portion of the C-130 crew force, initialairdrop and short field landing formal training (also known as "assault landing"training) is normally conducted at Little Rock Air Force Base, but class availabilitysometimes necessitates in-unit training at home station. Most C-130 special missionqualification training is conducted in-unit or at Kirtland Air Force Base (for specialoperations units and rescue).

3.4.1.3 Continuation Flying Training. Once crewmembers complete initialqualification and mission qualification, they must maintain that qualification bysuccessfully completing certain training events on a recurring monthly, quarterly, orsemi-annual basis. Referred to as "maintaining currency," accomplishing thesetraining events assures that the individual refreshes skills and familiarity withessential maneuvers, therefore staying qualified to fly the aircraft. This training isconducted in-unit.

3.4.1.4 Continuation Ground Training. Staying qualified to fly requirescompletion of other recurring monthly, quarterly, semi-annual, or annual ground trainingevents. These events cover a broad spectrum of flying-related activities to includephysiological training, annual flight simulator refresher training (in normal andemergency procedures), life support training, and cockpit resource management training(CRM). CRM training focuses on enhancing crew synergy, coordinating on missionaccomplishment and handling unusual emergency situations. Continuation classroom andsimulator training is usually more advanced and tailored to the experience of the crew.This training is usually conducted at home station, or at a remote satellite simulatorlocation. The focus of this training is on more in-depth systems knowledge and handling ofmore complex emergency procedures. This simulator refresher training normally lasts threedays with four hours of academics and four hours in the simulator each day.

3.4.1.5 Initial Water Survival Training. Currently course number SV-90-A,Non-parachuting Water Survival, is the required course for large aircraft (non-ejectionseat) crewmembers. The course focuses on ditching procedures to the exclusion of bailout.The BAR is concerned about two issues. First, it leaves the crew deficient in techniquesof over-water bailout. Second, it encourages the belief that ditching is the preferredoption. AMC/DOT is working an initiative to restore overwater bailout to the curriculum ofSV-90-A.

3.4.2 Contractor Aircrew Training System The Little Rock Air Force Base C-130classroom and simulator training (both formal school and continuation training) arenormally conducted by a civilian contractor (Hughes). Their instructor corps consistsmostly of retired military C-130 crewmembers with many years of C-130 flying experience.Additionally, the aircrew training system (ATS) contract requires the syllabus andinstructors to be responsive to the latest changes and safety issues in the C-130.Specifically, the contractor is required to include information from safety supplements intheir classroom training within 48 hours of safety supplement release. The contractor willalso update the simulator scenarios to include these safety issues. The intent is toexpeditiously raise crew awareness and train the force on the latest operational andsafety issues affecting the C-130.

3.4.3 Comments on Continuation Training Overall, the team found C-130initial and continuation training to be a thorough, effective and responsive process. Whena new safety issue is addressed in a safety supplement, incorporating it into thesimulator refresher classroom training within 48 hours, and ultimately incorporating itinto training scenarios, is an effective training tool. However, the BAR found severalissues that warrant comment, and which may have possible impact on C-130 training andoperations:

3.4.3.1 Wide Variety of C-130 Missions and Models. Formal school training is conductedon C-130E models. While these E-models have served as a good training platform for manyyears, modifications (EC, HC, WC, etc.) and modernization (H-1, H-2, and H-3 aircraft)necessitate that many students undergo differences training in their particular aircraftupon arrival at home station. The burden of teaching the operation of C-130 aircraft otherthan E-models has been delegated to the individual unit. The number of units and personnelneeding differences training grows with each new modification and modernization to theC-130. This philosophy is not consistent with an integrated training program that strivesfor economy of instruction and assurance of effective, standardized instruction. The teamis concerned that procedures and techniques taught in-unit may not be standardized, norreceive the degree of review and scrutiny normally associated with "formal"C-130 training.

3.4.3.2 Formal School Challenges. In the past few years, the formal school at LittleRock has had difficulty keeping up with the rising training demands (i.e., the number ofstudents to be trained). This increase in training demand has been due to a number offactors, to include: increased crew ratios for pilots and loadmasters, the "bankedpilot" program (pilots trained through Undergraduate Pilot Training but delayed inbeing assigned to an operational aircraft due to previous surpluses in the pilot force),increased numbers of first assignment instructor pilots (or "FAIPs") assigned toC-130s, and a shortage of available C-130 instructor pilots. The result is a heavilyworked training system where most training is accomplished at Little Rock. However, somestudents’ assignments are delayed while they await training, or they must accomplishthe training at home station with in-unit waivers. The goal should remain to train allinitial qualification training students at the formal school to promote standardization inoperating the C-130 worldwide.

3.4.3.3 Training Video. The team viewed a training video that was developed following aColombian Air Force ditching of a C-130. The BAR was in general agreement that this videowas an effective training tool. An updated training video would incorporate the lessonslearned from King 56, the Colombian ditching, the gunship mishap in Africa, and otherditching events, and would be a valuable training tool.

3.4.3.4 Ditching vs. Bailout Training. The BAR reviewed flight manual guidance forditching and bailout. T.O. 1C-130H-1 pg. 3-64, in "Bailout Over-water" states:

"Consideration of various unfavorable factors involved in an overwater bailout limits the decision recommending overwater bailout to several specific instances: namely, when visual contact is made with land, or adequate surface help; when wind and sea conditions are such as to preclude ditching; when fire or loss of control makes ditching impossible."

3.4.3.5 This discussion clearly favors ditching over bailout in over-water situations.The BAR believes this priority needs to be reviewed and revalidated in light of availableditching information and anticipated life support / survival equipment availability. TheBAR further found that there is currently no requirement to review ditching or bailoutdrills.

3.4.3.6 In our visits with crew members around the country, the BAR found them to befamiliar with ditching procedures outlined in their respective dash ones. The overwhelmingmajority were familiar with the general details of the Colombian aircraft ditching. Therewas an awareness among the crews that ditching, even under the best of circumstances,carried the probability of extensive damage to, and immediate flooding of, the aircraftshortly after initial impact. They acknowledged a strong tendency to want to stay with theaircraft, rather than bailout and be strung out across the water--away from the majorityof the available survival gear stowed aboard the airplane. The BAR believes the flightmanuals should be updated for the best information available on ditching and how toprepare for and successfully execute the maneuver. An expanded discussion of the merits ofditching versus bailout in the dash one would be helpful as well. Clearly, the crewsunderstand ditching to be an emergency procedure--the probability of success is directlyproportional to conditions at the time of ditching: daylight, calm sea state, lightweight,powered flight, & favorable wind direction. Under any other conditions, theprobability of success becomes marginal at best.

3.4.3.7 Related Deficiencies and Concerns

DEFICIENCY: Periodic review of ditching and bailout procedures by aircrews isnot currently required.

RECOMMENDATION: The BAR recommends that the Air Force establish a requirement for all crews to review ditching and bailout procedures on the first leg of overwater missions.

DEFICIENCY: Ditching information in flight manuals is inconsistent andinaccurate.

RECOMMENDATION: The BAR recommends that Air Force work to standardize this information between models of the same aircraft and reverify the accuracy of the information contained in the flight manuals on bailout and ditching, including the advisability of each, considerations involved in the decision, and the probability of survival in each case.

DEFICIENCY: Bailout over water information in the flight manuals needs to bereviewed. By favoring ditching over bailout, flight manuals are endorsing a recognizedprocedure with a low probability of survival. While this may be appropriate, the BARbelieves a review is warranted to include proposed survival equipment changes. Clearly,with passengers on board (depending upon their number, ages, experience, sea state, andavailability of parachutes and life support equipment), ditching may well be the onlyoption.

RECOMMENDATION: The BAR recommends that Air Force work to standardize this information between models of the same aircraft and reverify the accuracy of the information contained in the flight manuals on bailout and ditching, including the advisability of each, considerations involved in the decision, and the probability of survival in each case.

3.4.3.8 Systems Training. The BAR discovered an inaccurate belief within the crew forcethat the synchrophaser is responsible for most, if not all, power anomalies in thisaircraft. This "synchrophaser psychosis" results in both the crew force andmaintainers being spring loaded to blaming synchrophasers when other systems such asaircraft fuel, electrical, or bleed air might be involved. The BAR identified severalincidents where focusing on the synchrophaser delayed identifying fuel problems.

3.4.3.8.1 A critical action "bold face" procedure was developed to addressthe serious situation of multiple engine power loss/RPM rollback emergencies. Results ofBAR initiated flight tests have confirmed the inability of the engines to sustaincombustion due to fuel starvation in certain situations. These situations can result inpower losses that can be recovered by turning on the main fuel tank boost pumps andclosing the crossfeed valves. If no corrective action is taken all four engines mayflameout.

3.4.3.8.2 The flight manual has contained a four-engine rollback procedure for a numberof years. While not a bold face item, most, if not all units, expected crew members tocommit this procedure to memory, just as they would a bold face procedure.

3.4.3.8.3 This newly issued Multiple Engine Power Loss/RPM Rollback procedure has foursteps that are "bold face" (critical actions which must be committed to memory).The first two addresses fuel and ensures the fuel pumps are on and the fuel system is in"tank to engine" configuration. The third directs placing all propeller governorcontrol switches to "mechanical governing" This should remove all electricalinputs from the synchrophaser to the propeller control assembly. The final "boldface" step directs placing all temperature datum control switches to"null". This removes any electrical corrections from the TD system and returnsthe TD valve to a 20% bypass position. The first four steps of the procedure will allowthe crew to recover from either a possible fuel starvation situation or electricallyinduced synchrophaser or TD system malfunction. Once the situation is stabilized, and theaircraft is in its most basic operational mode, the crew will have time to analyze thespecific situation/malfunction.

3.4.3.8.4 The implications of the flight test results impact several procedures in theflight manual. For example, before the generators are turned off in the case of anelectrical fire, the flight manual should be updated to address closing the crossfeedvalves. This is only one example of impacts on the flight manual.

3.4.3.9 Cockpit Resource Management: Cockpit Resource Management has been anelement of aircrew training for several years. Despite this training both in the classroom and in the simulator, there continues to be mishaps where CRM is clearly asignificant factor. Currently, annual ground training is focused on improving crew synergyand coordination to enhance safe, efficient mission accomplishment in handling aircraftemergencies. The classroom portion of the training addresses human interaction,communications and group dynamics. Simulator training includes only those crew members whohave positions in the simulators and is not done by all commands. The concept is to applyclassroom lessons in scenarios in the cockpit. CRM training is conceptually good andprofessionally presented; however, anecdotal evidence suggests that the lessons are notfully incorporated into crew behavior. It is clear to the BAR that the CRM program needsto be reviewed. If the data is available, review a random subset of pre and post CRMtraining mishaps. Has the training improved the mishap rate? Is CRM properly focused, orcould modifications to the program improve the crews performance without adding to thetraining burden

3.4.3.10 Multiple Aircraft Configuration: There is considerable discussion within thisdocument concerning various aspects of configuration control. This issue has operationalimpacts. During their unit visits, the BAR found that several units had more than oneconfiguration of the same series of aircraft. This creates situations in which instrumentlocations, procedures and systems details are different.

3.4.3.11 Techniques vs. Procedures: A core principle in flying, both commercial andmilitary, is the strict adherence to established procedures. It appears that small,isolated flying communities have the potential to develop, intentionally or accidentally,techniques masquerading as procedures that have not been evaluated and approved byappropriate authority. One specific example, discussed in detail elsewhere, is thetechnique of turning off the main tank fuel boost pumps when feeding all four engines fromthe fuselage tank. The BAR discussed at length mechanisms to re-establish the sanctity ofthe flight manual, but that requires the repair of the flight manual system and theunification of the flight manuals under a single manager.

3.4.3.12 Aircrew Experience Level: Each command designates their standards foraircrew experience. The BAR reviewed crew force experience levels and found most operatingcommands met or exceeded those standards. Overseas, Special Operations, and both Guard andReserve units generally exceeded experience objectives by a large margin. These commandsalso exceeded the experience levels generally found in active duty CONUS combat aerialdelivery units. While important to monitor, the current crew force has a realisticcapability to meet current taskings safely.

3.4.3.13 Related Deficiencies and Concerns

DEFICIENCY: The BAR directed flight test identified flight manual discrepancieswith respect to the fuel management in certain situations.

RECOMMENDATION: The flight manual should be reviewed to incorporate the lessonsof the flight tests in relevant areas of the flight manual.

Section 4.0

Maintenance

4.1 C-130 Maintenance Training

4.1.1 The BAR sought to determine whether maintenance personnel receive adequatetraining to enable them to safely maintain C-130 aircraft. They examined the safetyaspects of the issue by: visiting sixteen different C-130 units, interviewing maintenanceand operations personnel and soliciting their views on maintenance training issues,reviewing incident reports looking for trends, listening to inputs from a toll-free hotline, and talking to depot technicians to get their views on the condition of hardwarereturned to the depot for repair. As a result of our investigations, two areas wereselected for special consideration:

4.1.2 On-the-Job Training (OJT): Since assuming lead command responsibilitiesfor C-130s, Air Mobility Command has continued to monitor and improve training programs.C-130 maintenance training had previously adopted a structured training program, calledOJT. This approach enables highly skilled craftsmen and supervisory-level Air Forcemaintainers, or their civilian counterparts, to train and certify new Air Force personnelin the hundreds of maintenance actions required to maintain Air Force aircraft. As aresult of this one-on-one approach, entering technicians learned their skills under thewatchful eyes of experienced veterans in their respective fields. The MaintenanceQualification Training Program (MQTP) standardizes OJT for each type aircraft throughoutthe command. The priority for implementing the MQTP concept was to first optimize theprogram on the C-141, C-5, KC-135 and C-17, before tackling the more difficult C-130 withits multiple mission design series (MDS) and configurations. This program has beensuccessful and the C-130 MQTP classes will begin in March 1998.

4.1.3 Training Program Improvements: AMC has also continued efforts to upgradetraining center mock-ups and trainers, and to develop computer based training (CBT). TheBAR found several areas that were noteworthy and other areas where the command was workingto improve maintenance training.

4.1.4 Summary: Inputs were received from AMC, ACC, AFSOC, AFRC, and the ANG. Theteam did not find any maintenance training issues which have flight safety implications.

4.2 Maintenance Experience Levels

4.2.1 The BAR reviewed maintenance personnel experience levels in all of theoperating commands. As expected because of their generally longer periods of service,individuals in the Guard and Reserve units generally exceeded the experience objectives,and exceeded the experience levels found in the active duty units. Although there has beena decrease in experience levels as the forces have been downsized, the team found noevidence that the decreased experience level is having an adverse affect on flight safety.

4.3 C-130 Maintenance Inspections

4.3.1 All Air Force aircraft are inspected before flight. For the C-130,aircraft inspections range in detail from the most common flightline inspections performedby aircrews immediately before flight, through all the maintenance inspections to readythe aircraft for the aircrew, all the way to a full programmed depot maintenance, or"PDM" inspection at the depot. Basic preflight and postflight inspections, homestation checks, and the isochronal (calendar-based) inspection processes are usually doneat the aircraft’s home base of assignment and are performed by the Air Forcemaintenance personnel assigned to that unit. The PDM inspection process goes on at thelarger depot facilities which are set up to handle the major maintenance associated withthis level of inspection effort. Each inspection is designed to look at only certainitems, which cuts down on duplication of effort between inspections. Items inspectedduring a home station check may not be looked at during the Isochronal inspections (called"ISO" for short) or might be inspected to a different level.

4.3.2 All of these inspection efforts are directed and governed by USAFtechnical orders, or T.O.s, which will be explained in Section 4.6. The total number ofinspection tasks that are performed on each aircraft in the fleet on a periodic basis isvery large, consuming a large amount of man-hours, in order to provide aircrews withaircraft that are safe to fly. It is common practice for some minor inspections to becomeoverdue and completed at the next scheduled maintenance event. This is authorized to allowflexibility in managing the flying schedule plus increasing aircraft availability. The BARdid not find any safety deficiencies or problems with the C-130 inspection process.

4.4 Depot Level Maintenance Impact On Flight Safety

4.4.1 The team posed the following questions: 1. Are there any flight safetyconcerns about depot level maintenance activities? 2. Are aircraft PDM and individual itemdepot overhaul activities producing aircraft and equipment that are safe to operate andmaintain? The team answered these questions on PDM by looking directly at the C-130fleet’s PDM, and its propeller system component overhaul processes, materials, andworkmanship, to identify any safety concerns or problems which might make the airplaneless safe to fly. They went directly to the depot facilities at Warner Robins AirLogistics Center in Robins, Georgia (WR-ALC) where the airplanes undergo major repair andtheir records are kept. The BAR looked especially hard at the C-130 aircraft fuel system,engine, and propeller systems. They found no significant safety issues with eitheraircraft or repaired parts coming out of depot level maintenance.

4.4.2 Aircraft PDM visits are based on calendar months since the last visit. C-130aircraft are scheduled for PDMs at specific intervals, depending on their mission. Duringeach visit to a PDM facility, the depot workers perform certain specific inspections,repairs, or refurbishment operations in accordance with a detailed PDM work package. Thedepot work package of repairs and inspections is agreed upon with the operational commands(e.g., Air Mobility Command, Air Force Reserve Command, and others) during the MaintenanceRequirements Review Board (MRRB) meeting held annually.

4.4.3 The depot also performs any "over and above repairs" (more than whatwas called for in the agreement) on specific tail numbers as requested by the operationalunit owning the aircraft. This might range from paint touchup work to prevent corrosion torepairs carried over as important but not critical enough to prevent safe flight.Depot-level tasks are those "heavy maintenance" tasks that call for moreexpertise, tools, and special heavy equipment than local flying units normally have. Inaddition, PDM involves inspections generated under the aircraft structural integrityprogram (ASIP). The ASIP is an extremely rigorous process, usually involving the originalaircraft manufacturer, to ensure that the model of aircraft in question does not suffer acatastrophic structural failure. This is accomplished by performing specific structuralinspections, repairs, and replacements developed by engineering analysis, individualaircraft usage monitoring, and in some cases, data obtained by full-scale fatigue testing(i.e., bending and vibrating a part over and over to simulate thousands of hours ofoperational service to see what happens in the laboratory instead of in flight.) Thisarrangement is beneficial because the equipment for large-scale repairs and fatiguetesting is costly and does not need to exist at every location.

4.4.4 From the safety perspective, the key questions relating to the PDM process are:1. Are the right inspections and repairs being accomplished, i.e., is the MRRB processcorrectly identifying the work that needs to be done, and 2. Is the agreed upon work beingperformed correctly during PDM? The Air Force’s experience with this process over thelast several decades has proven the MRRB process to be effective in ensuring that thenecessary safety-related depot level maintenance is identified and being performed.

4.4.5 The primary measures of success for PDM quality from the safety perspective isthe number and significance of deficiencies reported by operational units upon receipt ofthe aircraft after its PDM visit. Team members reviewed the past two years of this dataand found no safety-related problems. Although field units have expressed some concernfrom time to time over such issues as the quality of paint application on PDM aircraft andvarious minor workmanship defects, the overall quality levels are satisfactory. Reporteddefects were corrected or resolved to the satisfaction of the operational units flying theairplanes.

4.4.6 Component Overhaul. The Air Force has overhauled aircraft parts for manyyears. C-130 parts are overhauled under the Management of Items Subject to Repair, orMISTR, program. MISTR overhauls items "on-condition," i.e. when the item nolonger performs satisfactorily.

4.4.7 The team visited the propeller and the synchrophaser overhaul facilities,interviewed workers, and reviewed production quality records. The BAR looked at theresults of investigations into quality deficiencies as well. A review of records revealthat over 90% of all synchrophasers returned for repair had no deficiencies. A samplereview of 48 synchrophasers returned for repair revealed no "repeat" offenders.The BAR found no safety related deficiencies and judged these overhaul processes to besound and effective in producing safe, quality items. The BAR did find that five aircrafttail numbers had repeat synchrophaser problems. When these synchrophasers were tested,there was nothing wrong with them. This points to other systems on the aircraft adverselyaffecting the synchrophaser.

4.4.8 Related Deficiencies and Concerns

DEFICIENCY: The depot was previously unable to properly track propulsion systemcomponents.

ON-GOING RESOLUTION: The depot at Warner-Robins has initiated a serially numbered tracking program to obtain the necessary data in an effort to explain what other aircraft systems affect the synchrophaser.

4.5 Maintenance Practices

4.5.1 The BAR found concerned, qualified personnel who found and fixed the problemsreported by the flight crews, or who discovered and fixed problems themselves duringroutine inspections. However, the team saw two instances that are cause for concern.First, virtually no one in the field appeared to be sampling the fuel in aircraft fueltanks (referred to as "pogoing" the tank, based on the use of a long hollow pole["pogo stick"] with a collector jar at the end for visual inspection of the fuelsample). Uniform use of this procedure would more readily identify the presence of waterand other impurities in aircraft fuel tanks. Second, a core principle of aircraftmaintenance is being overlooked in some cases. The BAR found that some maintainers werenot consistent in their use of technical order trouble-shooting procedures and theassociated required follow-through maintenance actions. In particular, the team noticedthe same "synchrophaser psychosis" previously mentioned had a tendency to causemaintainers to immediately assume the synchrophaser was at fault, rather than tothoroughly test it using the appropriate test equipment and procedures. There were alsoindications that many maintainers were unfamiliar with how to properly test thesynchrophaser. This may well contribute to the trend, noted in Section 4.4 of this report,of 90% of synchrophasers sent to depot for repair showing no need for such repair.

4.5.2 Related Deficiencies and Concerns

DEFICIENCY: Maintainers are not sampling aircraft fuel as required by the T.O.

RECOMMENDATION: The BAR recommends renewed Air Force-wide emphasis on fuel sampling as part of standard maintenance operations to help identify the presence of water or other contaminants in the fuel.

DEFICIENCY: Maintainers are not always properly trouble-shooting reportedsynchrophaser malfunctions and may not be thoroughly familiar with the procedures requiredfor testing the synchrophaser.

RECOMMENDATION: The BAR recommends renewed Air Force-wide emphasis on proper testing of the C-130 synchrophaser following reported malfunctions, and on more thorough training for maintenance personnel in the performance of those tests.

4.6 Technical Orders

4.6.1 The team addressed the following questions on technical orders: 1. Are there anysafety concerns in the USAF Technical Order System? 2. Are all of the books, manuals, andchecklists used by personnel to maintain C-130 aircraft accurate and effective to ensurethat safety is being maintained?

4.6.2 There are detailed procedures to ensure that all of these T.O.s are kept up todate and that any deficiencies discovered are corrected. Any crewmember or maintenanceperson can identify a deficiency by writing it up on the proper form--Form 847 for flightmanuals and an AFTO Form 22 for maintenance manuals. These forms are submitted throughchannels to the System Program Office (SPO) who is charged with the responsibility formanaging these T.O.s.

4.6.3 The BAR looked very closely at this process to ensure that all such deficiencieswere being addressed in a timely manner to preserve flight safety. They found nounresolved deficiencies in this process, both for flight manuals and maintenance manuals.However, the same issues that exist with aircrew flight manuals exist with maintenancetechnical orders as well: a large backlog of changes to produce and post, multiple changepages to search through to accomplish even relatively simple maintenance actions, andscarce funding to solve the problem. The BAR is concerned that, while the veteran linecraftsman may know where to look for all the changes when completing a repair, the lessexperienced maintainer may miss a critical step that is buried in a series of supplements,with potentially serious consequences. It will require a significant investment inresources and time, over $20 million and approximately two years, to fix the C-130 aloneusing current manpower levels to correct.

4.6.4 Initiatives are also underway to convert USAF technical manuals from the old,expensive and time-consuming paper format to the newer digital format. New CD-ROMtechnology offers many benefits, including a reduction in the annual $2.5 million cost ofmaintaining our T.O.s. This conversion faces many obstacles, including the cost ofconversion as well as training and equipping field units to handle electronic data ratherthan paper.

4.6.5 Related Deficiencies and Concerns

DEFICIENCY: Maintenance units do not sump fuselage tanks on a regular basisleading to the possibility that water and contamination could collect within these tanks.

ON-GOING RESOLUTION: The BAR supports amendment of the T.O.s as necessary so that fuselage tank sumping is required at regular intervals.

Section 5.0

C-130 Mishap Review

5.1 USAF Mishap Data

5.1.1 The USAF C-130 has a very strong safety record. Introduced into the AirForce inventory in 1955, the C-130 has amassed over 14,400,000 flying hours. During thistime, the Air Force experienced 142 Class A mishaps (aircraft destroyed or damaged beyond$1 million or economical repair, or where permanent disabling injury or loss of lifeoccurs) resulting in 613 fatalities and the loss of 83 aircraft. An additional 45 aircraftwere lost to combat. Since 1971, the Air Force has experienced 63 C-130 Class A mishapsresulting in the loss of 54 aircraft.

5.1.2 The C-130 has followed the USAF trend of fewer mishaps per flying hour over theyears. The BAR attributes this to a number of factors including, but not limited to,increased systems reliability, improved components, improved training, and the USAF safetyprogram.

5.1.3 The C-130 program implements the USAF safety program in two ways. The SystemSafety Group consists of all C-130 users and is focused on mishap prevention. The MaterialSafety Task Group tracks the status of all appropriate USAF C-130 mishap recommendationsand ensures the appropriate resources are applied and progress is being made on thecorrective actions resulting from these recommendations. Safety is also integral to C-130training and the content of the technical orders.

Figure 5-1

USAF Historical Mishap Vs. C-130 Mishap Rates:

1957-1997

C-130 Broad Area Review (7)

5.1.4 The cumulative class A mishap rate for the C-130 is 0.99 (class A mishaps per100,000 flying hours). This safety record is noteworthy when considering the missions andenvironments in which the USAF flies the C-130 (see Section I, Operating Environments andMissions). The C-130 rate is well below the Air Force rate of 1.37 and comparable to theC-5 rate of 0.91.

5.2 Worldwide Mishap Data

5.2.1 The U.S. Navy and U.S. Coast Guard also fly C-130 aircraft. The Navy has flownC-130s since 1961. Their lifetime class A mishap rate (1961-1998) is 0.87 mishaps per100,000 flying hours, but they’ve had zero class A mishaps since 1977. The CoastGuard only has flying hours available back to 1983. From 1983 to 1997, their class Amishap rate is 0.30 (only one mishap). Between 1961 and 1982 they experienced three otherclass A mishaps.

5.2.2 With almost 25 million C-130 flying hours world wide, there have been 284aircraft lost to mishaps: 194 Class A mishaps, 14 ground mishaps, four other mishaps, and72 lost in combat. Data on the causes of these mishaps (not including combat losses) isdepicted in Figure 5-2. Of the approximately 2,100 aircraft built in the last 44 years,approximately 1,800 are still in service.

Figure 5-2

C-130 Broad Area Review (8)

5.3 Uncommanded Power Reductions

5.3.1 Analysis of Reported Uncommanded Power Reductions. Table 5-1 shows theknown reported incidents of uncommanded power reduction since the Air Force began keepingthese records in 1983. Note that none of these are Class A mishaps, and that the list doesnot include the Portland King 56 mishap.

Table 5-1

Breakdown of 71 Reported Incidents:

Electromagnetic Interference (EMI) 03

Fuel Starvation 03

Synchrophaser 07

Aircraft Electrical System 24

Unknown 34

5.3.1.1 Electromagnetic Interference (EMI): EMI from the HF radio antenna accounted forthree of 71, or 4% of the reported incidents. One incident resulted from an improperlyconnected HF antenna. Shielding has worked in keeping the number of electromagneticinterference incidents down.

5.3.1.2 Fuel Starvation: During the course of reviewing the reported C-130 power-lossincidents, three events were of special interest due to their apparent similarity to thePortland mishap. These events were clearly sequential engine power-loss events, not thetraditional simultaneous engine power-losses historically associated with synchrophaser orelectrically related power-loss events. It was postulated that these events were reallyfuel starvation events and an effort was made to learn more about them. Additionally, noneof these events resulted in a mishap so the corrective actions taken by the crews werealso of interest since the actions may help improve existing procedures. To learn moreabout each of these incidents, the team contacted the flight crews for each incident. Inthe process of examining the details surrounding these three events, another unreportedevent was discovered which also exhibited the symptoms of fuel starvation. This event wasalso examined, bringing the total looked at to four. A discussion of each of theseincidents is contained in Section 5.4.

5.3.1.3 Synchrophaser: The synchrophaser accounted for only seven of 71 or 10%of the reported incidents. Of these seven incidents, three were due to water getting intothe synchrophaser unit, two were due to interface wiring bundle problems, and two were dueto internal synchrophaser problems. This is consistent with the fact that 90% of thesynchrophasers returned to the depot from the field for deficiencies under the productquality deficiency report (PQDR) system tested within operational limits.

5.3.1.4 The PQDR is the unit’s way to get feedback from depot when they send adefective part in for repair. The unit requests PQDR action on a specific part byproviding the specifics of the malfunction to the depot. The depot analyzes and repairsthe part, then identifies in writing to the sending unit what they found. Thesynchrophasers examined by the depot were found to be within acceptable tolerance andadjusted back to centerline, then returned to the field. The small percentage of defectivesynchrophasers caused the team to look more toward other potential causes for enginerollback and other uncommanded power reduction phenomenon.

5.3.1.5 Aircraft electrical system - This system may have contributed to 24 of 71 or34% of the reported incidents. The electrical system’s components may have fed faultyor fluctuating power, or data signals, to the synchrophaser or to the electrical controlswithin the propellers themselves. Possible problem sources include the failure ofgenerators, generator control panels, and the essential AC bus. Faulty electrical systemcomponents may have also played a part in the majority of the 34 incidents with unknowncauses, making it a category of considerable interest. The ongoing FMECA should revealadditional information on what part the aircraft electrical system plays as a potentialcause of problems.

5.3.1.6 Unknown - Thirty-four of 71, or 48% of the reported incidents are classified ascaused by unknown reasons. Most of these are strongly suspected to be caused by theaircraft electrical system (old synchrophaser interface wiring bundles, bad grounds, oldpower system powering newer components, etc.).

Figure 5-3

C-130 Broad Area Review (9)

5.3.1.7 Since completing the installation of the solid state synchrophaser in June1992, there have been no reported rollbacks attributed to internal failure of thesynchrophaser and only four rollbacks with unknown causes. This indicates that efforts toclean up the electrical power and improve synchrophaser performance have been beneficial.Modifications include the constant voltage transformer (Dec 88 - Dec 93), solid statesynchrophaser (Jan 90 - Jun 92), and HF antenna lead shielding (Mar 92 - Mar 98 [est.]).The ongoing FMECA should identify any additional potential problem areas with the aircraftelectrical system.

Table 5-2

Breakdown of 71 Reported Incidents by Tail Number (Year ofManufacture):

55 - 01

56 - 03

57 - 02

58 - 00

59 - 00

60 - 00

61 - 05

62 - 03

63 - 08

64 - 12

65 - 07

66 - 02

68 - 05

69 - 05

70 - 01

72 - 00

73 - 03

74 - 12

78 - 00

79 - 00

80 - 00

81 - 00

82 - 00

83 - 00

84 - 00

85 - 01

86 - 00

87 - 00

88 - 00

89 - 00

90 - 00

91 - 00

92 - 01

93 - 00

94 - 00

95 - 00

96 - 00

(Note: Intervals between years reported [e.g. 75-77] reflect no C-130purchases by USAF)

5.3.1.8 Only two of the incidents reported thus far occurred on aircraft built after1974 (see Table 5-2). The majority of the reported incidents (63 of 71 or 89%) occurred onaircraft built and fielded between 1961 and 1974. The lack of incidents associated withnewer aircraft, coupled with the fact that there have been only two incidents caused byinternal failure of the solid state synchrophaser, combined to discount the solid statesynchrophaser as a likely cause of the problems experienced.

Table 5-3

Breakdown of 71 Reported Incidents by C-130 Mission Design Series:

C-130A 04

C-130B 03

C-130E 26

C-130E (ESU) 01

C-130H-1 12

C-130H-2 01

C-130H-3 01

AC-130A 01

AC-130H 02

DC-130A 01

EC-130H 04

HC-130H 04

HC-130N 02

HC-130P 04

MC-130E 03

WC-130H 02

5.3.1.9 Basic C-130 aircraft (A, B, E and H models) accounted for 68% or 48 of the 71incidents (see Table 5-3). These aircraft, however, comprise the overwhelming majority ofthe C-130 fleet (75.8% of the Air Force’s fleet, or 526 of 694 as of April 1, 1997).Our modified aircraft, with their additional systems installed, tend to have a higherpercentage of reported incidents of RPM rollback (see Figure 5-4).

Figure 5-4

Fleet Composition vs. Incidents

C-130 Broad Area Review (10)

5.4 Possible Fuel Starvation Incidents

5.4.1 Three of these 71 incidents, although initially reported as RPM rollbacks, weredetermined by the BAR to have occurred due to fuel starvation. Additionally, one otherunreported incident was also determined to be caused by fuel starvation. In each case,this determination was made based upon the incident report (if reported), crew testimony,and system analysis. Each of these incidents is discussed below.

5.4.2 Spring 1991 Incident, HC-130N. This aircraft departed home station in themorning and refueled three helicopters. It landed at a second air field, refueled, andprepared to refuel helicopters again. In the afternoon, it refueled three helicopters in ashort amount of time. During this swift refueling operation, the aircraft ended up in a"secondary fuel management" condition. The flight engineer accepted thisposition as a cost of being able to offload fuel to receivers rapidly. However, whenrefueling operations were complete, the flight engineer worked to get back into a primaryfuel management position. This was accomplished by "scavenging" small amounts offuel remaining in the auxiliary and external tanks, using the remaining fuel in thefuselage tank, and balancing main tank fuel with two main tank pumps on and the other twooff. According to testimony, primary fuel management was achieved, all main tank pumpsturned on, and the aircraft fuel panel in tank-to-engine configuration prior to theaircraft entering a low-level route for its return to home station.

5.4.2.1 After a short period of flight in the low-level route with light turbulence,one of the torque gauges began to fluctuate slightly. After confirming that this enginewas not the master engine for the synchrophaser, the torque on number 3 engine gauge wasobserved to drop significantly. The remaining engines also began to lose power as well.Nearly simultaneously with the power-loss, the aircraft commander took control of theaircraft, initiated a climb, and the flight engineer began to execute the four-enginepower-loss procedure.

5.4.2.2 While the flight engineer was executing the four-engine power-loss procedure,the load master stated, over the intercom, that the number 1 engine had flamed out. Theaircraft commander visually scanned the number 1 engine and confirmed its condition. Theflight engineer, while on the floor by the pilot’s seat preparing to pull thesynchrophaser’s AC circuit breaker, also confirmed the condition of the number 1engine. The copilot, reacting to the load master’s observation and the pilot’sverification was ready to feather the number 1 engine and was awaiting confirmation fromthe flight engineer before doing so. The flight engineer returned to his seat withoutpulling the synchrophaser’s DC circuit breaker. Confirmation to feather the number 1engine was not given to the copilot because it appeared that power was beginning to returnto the other engines. Power did return to number 2, number 3 and number 4 engines andnumber 1 started as if the crew were performing an air start. All engine power wascompletely restored.

5.4.2.3 The crew diverted to a nearby airfield without further incident. Specificmaintenance actions performed on the aircraft and the results of fuel samples taken fromthe aircraft and the three helicopters are unknown. This aircraft was equipped with atube-type synchrophaser.

5.4.2.4 According to the testimony provided the BAR, during the period of the rollback,all four low fuel pressure lights were illuminated, as were the number 1, number 3 andnumber 4 engine generator-out lights. With the number 2 generator turned off (a result ofexecuting the four-engine power-loss procedure), and the other generator-out lightsilluminated, there should have been no electrical power being generated by this airplane.The BAR knows this cannot be true, otherwise the aircraft low fuel pressure lights (whichreceive power from the DC essential bus), would not have been illuminated. Since the lowfuel pressure lights were illuminated, at least one generator must have been on-line andproducing power. With only one generator on-line, the number 2 main tank pump (whichreceives power from the essential AC bus), and the number 3 main tank pump (which receivespower from the Main AC bus) must have been running--provided they were turned on. If theyhad been turned on and running, they would have been producing pressure, thusextinguishing the number 2 and number 3 low fuel pressure lights. Since the number 2 andnumber 3 low fuel pressure lights were illuminated, the BAR believes that the number 2 andnumber 3 main tank pumps were turned off. With the other engines behaving similarly, theBAR also believes that those main tank fuel pumps were turned off as well.

5.4.2.5 This last conclusion is based upon evidence that, upon the flight engineerreturning to his seat, all four main tank boost pumps were cycled off and then back on. Ifthere had been an electrically related problem, it is doubtful that this action would havecorrected the situation. The BAR believes that the main tank fuel pump switches were offand that by "cycling" them, they were actually turned on, providing a positiveflow of fuel to the engines, resulting in the restoration of all engine power. It isbelieved that instead of the two main tank pumps being turned back on prior to enteringthe low-level route, the two that were already on were actually turned off. Even thoughthis aircraft was flying a low-level route, it is believed that pressurized cabin airentered through an empty fuselage tank and worked its way out to the engines. This couldhave happened, provided the low-level route was being performed over high terrain andcabin pressure was set at a typical value of 1,000 feet.

5.4.2.6 With this air in the manifold, it eventually worked its way out to all fourengines resulting in the engine power-losses. It is believed that this air has thepotential to adversely affect critical engine fuel control components, disturb/disruptcombustion within the engines, or cavitate the engine-driven fuel boost pumps. By turningthe main tanks pumps on, it is believed that the source of the air was eliminated, andonce the existing air was purged from the fuel supply manifolds, power to the engines wasrestored. This aircraft did not experience any further problems similar to what this crewexperienced.

5.4.3 Fall 1992 Incident, HC-130P. This incident occurred during an aircraftferry mission, flown to return the aircraft to home station. While cruising atapproximately FL 180 (18,000 feet), one hour into the sortie, the pilot noticed torque,fuel flow and TIT start to decrease on the number 1 engine. A short while later, thenumber 3 engine exhibited similar problems. Then number 2 and number 4 engines started tolose power as well. The flight engineer initiated the four-engine power-loss procedure. Asthe final steps of the procedure were accomplished, the following results were obtained:pulling the synchrophaser’s AC circuit breaker resulted in no affect while pullingthe synchrophaser’s DC circuit breaker apparently restored power to all engines. Thecrew then flew one hour back to home station, the nearest field, without further incident.

5.4.3.1 Maintenance could not find anything definitively wrong with this aircraft butremoved and replaced the aircraft’s solid state synchrophaser followed by uneventfulengine runs. The same crew then flew this aircraft without further problems.

5.4.3.2 Testimony provided the BAR revealed that the flight engineer’s personaltechnique of turning off the main tank fuel pumps was utilized when feeding all fourengines from a fuselage tank. In this case, the crossfeed separation valve was open, allfour engines were being fed from the fuselage tank, and all the main tank pumps wereturned off. After the fuselage tank empty light had flickered for a few seconds, thesecond pump in the fuselage tank was turned on and allowed to run briefly before theaircraft was transitioned to crossfeed from the external tanks. To accomplish this, theexternal tank pumps were turned on, the fuselage tank pumps turned off, the crossfeedseparation valve was closed, and the main tank pumps turned on. While this was beingaccomplished, all four engines began to lose power as described above.

5.4.3.3 The BAR believes that, for the period of time between the fuselage tank emptylight first flickering and turning on the external tank pumps, pressurized cabin airentered the fuel supply manifold via the empty fuselage tank. With this air in themanifold, it eventually worked its way out to all four engines resulting in the enginepower-losses. It is believed that this air has the potential to adversely affect criticalengine fuel control components, disturb/disrupt combustion within the engines, or cavitatethe engine-driven fuel boost pumps. By transitioning to a new source of fuel (i.e., theexternal tanks), it is believed that the source of the air was eliminated, and once theexisting air was purged from the fuel supply manifolds, power to the engines was restored.Lastly, it is believed that pulling the synchrophaser’s DC circuit breaker and thereturn of engine power coincidentally correspond to the time when the last of the airfinally worked its way out of the crossfeed manifold. This aircraft did not experience anyfurther problems similar to what this crew experienced.

5.4.4 Winter 1992 Incident, HC-130P. This incident occurred during a ferrymission. While cruising somewhere between FL220 and FL250 all the engines began tosequentially lose power. The exact order of the power losses could not be determined buttorque losses of approximately 2,000-3,000 in-lbs were seen on all the engines. The crewdid not run any portion of the four-engine power-loss procedure. Instead, they came offcrossfeed and went tank-to-engine and all engine power was restored. Just prior to enginepower-loss, the crew had been crossfeeding the fuselage tank to all four engines. Thefuselage tank was also running very low on fuel. The crew suspected that they had fuelcontamination, but results of fuel testing and fuel filter checks are not known.

5.4.4.1 In this incident, it is believed that the main tank pumps were off and when thefuselage tank ran low on fuel, pressurized cabin air was allowed to enter the fuel supplymanifold via the empty tank. With this air in the manifold, it eventually worked its wayout to all four engines resulting in the engine power-losses. It is believed that this airhas the potential to adversely affect critical engine fuel control components,disturb/disrupt combustion within the engines, or cavitate engine-driven fuel boost pumps.By returning to tank-to-engine configuration, it is believed that the source of the airwas eliminated, and once it was purged from the engine feed lines, power to the engineswas restored.

5.4.5 Summer 1997 Incident, C-130H. This incident occurred during the climbportion of a cross-country sortie. The aircraft was climbing through FL220 headed forFL240. The aircraft was being crossfed from the external tanks. Specifically, theleft-hand external tank was feeding number 1 and number 2 engines while the right-handexternal tank was feeding number 3 and number 4 engines. The crossfeed separation valvewas closed. While passing through FL220, small torque fluctuations were observed on thenumber 1 engine. This quickly progressed into larger power fluctuations on the remainingengines. The crew observed fluctuating torque, TIT and fuel flow on all these engines. Theaircraft was leveled off, and the flight engineer, knowing the Portland mishap happenedonly months before, began executing the four-engine power-loss procedure. The flightengineer never completed the procedure--neither synchrophaser AC nor DC circuit breakerwas ever pulled. Instead, the flight engineer came off crossfeed, went tank-to-engine, andengine power was restored.

5.4.5.1 The external tanks were low on fuel, indicating approximately 200 lbs each,just prior to engine power being lost. After restoring engine power, the aircraft wasflown for several more hours without incident. Fuel contamination in the external tankswas suspected by the crew, but feeding one engine exclusively from one external tank onthe ground for several minutes did not result in any anomalies. The external tanks werealso "sumped" in order to check for contamination. None was observed. Theaircraft was refueled and the cross-country continued without further incident. In total,three sorties were flown by the same crew after the power-loss event and no engineanomalies were noted during any of these subsequent sorties.

5.4.5.2 In this incident, it is believed that the main tank pumps were off and when theexternal tanks ran low on fuel, air entered the fuel supply manifold and eventually workedits way out to the engines. It is believed that air has the potential to adversely affectcritical engine fuel control components, disturb/disrupt combustion within the engines, orcavitate engine-driven fuel boost pumps. By returning to tank-to-engine configuration, itis believed that the source of the air was eliminated, and once it was purged from theengine feed lines, power to the engines was restored.

5.4.5.3 The BAR’s belief that main tank pumps were off is supported by testimonythat the flight engineer used a personal technique on the ground prior to flying localsorties (where external tank fuel is not required to complete the sortie), which was to"dry-drain" the external tanks. This technique involved feeding the engines fromthe external tanks, with the main tank pumps turned off, until all the fuel in theexternal tanks was exhausted. With this ground technique established and performed on theground, it is believed that a habit pattern was formed, and the personal technique waseventually utilized in flight and led to the incident just described.

Section 6.0

King 56 Accident Summary

6.1 Established Facts

6.1.1 The accident investigation report provides the summary of known facts surroundingthe crash of HC-130P, tail number 64-14856 and the death of 10 of the 11 people on board.This discussion will focus only on the aircraft and those operational and logistics issuesthat could potentially cause four engines to fail.

6.2 Flight Operations

6.2.1 Based upon digital flight data recorded (DFDR) information, the mishap aircraftdeparted Portland IAP at 1720 PST on 22 Nov 96 on an instrument flight rules (IFR) flighten route to North Island Naval Air Station. The purpose of the sortie was to conduct anoverwater navigation evaluation. King 56 began the sortie with a normal takeoff, departureand climbout. One hour and 24 minutes after takeoff in level flight at FL 220 the mishapsequence began with the engineer commenting on a torque flux on the number 1 engine.Nothing on the cockpit voice recorder (CVR), the DFDR, or the survivor’s testimonysuggested any unusual events prior to the engineer’s comment. Over the next threeminutes, the operations of all four engines became unstable and eventually failed. Crewactions during these critical three minutes are known only by verbal comments on the CVRand the survivor’s testimony. The following discusses what we know of those actions.

6.2.2 The engineer called for number 1 propeller to be placed in mechanical governing.This would normally remove electrical inputs to the propeller through the synchrophaser.The pilot then called for all four propellers to be placed in mechanical governing. Thisaction was consistent with treating this emergency as a four-engine rollback. There is noindication on the DFDR or the CVR as to whether or not the crew selected mechanicalgoverning on any of the remaining three propellers. At the same time the crew wasanalyzing the emergency, they also declared an in-flight emergency with Oakland ARTCC andturned the mission aircraft east to proceed toward Kingsley Field, Klamath Falls, OR,approximately 230 miles away and approximately 80 miles from the coast. The Radio Operatorradioed the USCG Humboldt Bay Station and notified them of the in-flight emergency. Duringthe turn toward the shore the number 3 and number 4 engines once briefly recovered most oftheir torque. These increases are recorded by the flight data recorder. When the RPM onnumber 3 (the aircraft’s last functioning engine) finally decreased below 94% RPM thelast generator producing electrical power dropped off line due to low frequencies. As aresult, at 1846 Pacific Standard Time all electrical power was lost. After a brief period,power was restored to the equipment powered by the battery bus. From this point on, theaircraft glided to the attempted ditching. There is no record of that portion of theflight, except the survivor’s testimony.

6.2.3 Other issues of significance. There are several extant records that document fuelinformation. The AFTO Form 151A indicates that King 56’s first sortie on 22 Nov 96landed with 12,000 lbs of fuel. The aircraft was then refueled. Fuel truck recordsindicate an onload of 4,088 gallons for the second sortie, which is approximately 27,800lbs. This data is consistent with an initial fuel load of 40,000 lbs on the mishap sortie.However, the Form F (AFTO Form 365-4) indicates a fuel load of 39,000 lbs. The fuel loadis normally documented in the aircraft records, AFTO Form 781, which were lost with theaircraft. The AFTO Form 781 is normally the principal place to record fuel distribution,so the BAR was unable to confirm the actual distribution of the initial fuel load. TheForm F is the aircraft weight and balance form completed before each flight and itnormally records the total fuel and its distribution. The Form F indicates that fuel wasin the main, auxiliary and external tanks.

6.2.4 Computerized maintenance records indicate that the right-hand auxiliary tank hada leak, and that the left-hand fuselage tank had an indicator problem, as did one externaltank. While unusual, this combination of inoperative gauges is permissible to fly inaccordance with regulations if certain procedures are followed. These restrictions wouldinclude not fueling the leaking tank, and in the case of the external tanks, lateralweight limitations require that the external tanks weigh the same amount. In this case,aircrews must verify the tanks full or empty. This would force the crew to either fuelthem completely full, or not fuel them at all. These are the only two ways to becompletely certain that the tanks weigh the same.

Table 6-1

King 56 Fuel Tank Malfunctions, Options, and Capacities

FUEL TANK MALFUNCTION FUELING FUEL CAPACITY

OPTIONS IN POUNDS

OUTBOARD MAINS* NONE STANDARD 6,834

INBOARD MAINS NONE STANDARD 7,322

LEFT AUXILIARY NONE STANDARD 5,598

RIGHT AUXILIARY LEAK NO FUEL 5,598

LEFT EXTERNAL INOP INDICATOR FULL OR EMPTY 8,359

RIGHT EXTERNAL NONE SAME AS LT EXT 8,359

LEFT FUSELAGE INOP INDICATOR NO FUEL 11,016

RIGHT FUSELAGE NONE STANDARD 11,016

*Refueling Pods Installed

6.2.5 The Form F and maintenance data conflict. The Form F indicates fuel in theexternal tanks, but not enough fuel to fill the tanks. The Form F also indicates no fuelin either fuselage tank.

6.2.6 Two possibilities exist: the Form F was either in error or correct. The BARexplored the first possibility. There was no indication that the external tank entry wasan error because the center of gravity calculations match those for external tanks shownon the Form F. Alternatively, the loadmaster could have anticipated one fuel load, butdiscovered another different fuel load had actually been put on the aircraft. Because thevariation in center of gravity was only one and a half percent from that expected, therewould have been little incentive to correct the error. Had the Form F been wrong, failureto correct the error would have had no safety implications. Finally, there exists thepossibility that the fuel was put in the external tanks, either by error, or because themaintenance records were in error and the external tanks were in fact operational. Whilewe cannot conclusively rule out either possibility, interviews with other engineers,loadmasters, and maintenance personnel caused the BAR to believe strongly that the fuelwas put into the fuselage tank and not the external tanks.

6.2.7 The mishap sortie was the second of the day for the aircraft. The mishap aircraftflew another training sortie earlier in the day. The pilot’s and radiooperator’s integrated display control unit (IDCU), a portion of the Self ContainedNavigation System, were replaced after the first sortie. Ultimately, the IDCUs weredetermined not to be defective. Discussions with maintenance personnel indicated a normalthru-flight was accomplished. No unusual maintenance was performed on the aircraft.

6.2.8 Two issues were raised about the mishap aircraft. The first concerned themicro-burst which struck the aircraft while parked on the Davis-Monthan AFB, AZ, ramp inMay 1994. The major mircoburst damaged the number 4 propeller, the right wing aft lowerspar cap, and the right inflight refueling pod and pylon. The number 4 engine andpropeller were replaced, as were the right refueling pod and pylon. The wing spar was alsorepaired. No documents were found to verify any additional inspections specificallyconducted with respect to this repair. It is appropriate to note that these repairs werecompleted in conjunction with programmed depot maintenance which was completed in July,1995.

6.2.9 The second issue, raised with both unit engineers and maintenance personnel, waswhether any maintenance was performed on all four engines or propellers on that aircraftin recent memory. That question generated two responses. There was a write-up of theaircraft pulling in one direction during taxi. After performing extensive maintenance, theproblem was ultimately resolved by the unit maintainers trading the number 1 and number 4propellers. This was done in September 1996.

6.3 Scenario Introduction

6.3.1 The team’s approach to understanding the King 56 mishap was to ask whatcircumstances could cause a C-130 to lose all four engines. A conscious effort was madenot to approach the issue sequentially, but to come up with as many theories as couldpotentially explain the mishap, however unlikely, and then evaluate scenarios based on theavailable data. The following is a general discussion of some critical aspects of aircraftoperations and the clues present in the King 56 data that point toward one scenario oranother. For each scenario presented, the BAR will offer both corroborative and rebuttingevidence as they apply.

6.3.2 Fuel Starvation: In its simplest form, fuel starvation is merely an inadequatesupply of fuel to the engines which lasts long enough for an engine to flameout. Fuelstarvation can occur if the fuel supply simply ceases to flow to the engines, issufficiently restricted, or if air or water is introduced into the fuel supply linesleading to the engines. Scenarios outlining how the cessation of, or restriction of, fuelflow, or the introduction of air or water into the fuel supply manifold could havehappened are discussed below. However, fuel starvation clues from the Portland mishap, andC-130 fuel system information needed to understand fuel-related scenarios, are onlypresented once.

6.3.3 Fuel Starvation Clues: There are at least five clues from the Portland mishapthat suggest fuel starvation. Prior to any hint of a problem surfacing, all four enginesare producing approximately 11,000 in-lbs of torque. The problem was first recognized at18:43:51 after torque on the number 1 engine had rapidly dropped to approximately 5,000in-lbs. Within 10-15 seconds, all four engines experienced varying degrees of torquereduction.

6.3.3.1 Clue One: At 18:44:02, the flight engineer stated that fuel flow to the number1 engine had "…gone to s---."

6.3.3.2 Clue Two: At 18:45:07, the flight engineer stated that he had lost fuel flow tothe number 2 engine. As the sequence continued to 18:45:10, the number 1 engine wasshutdown, the number 2 engine indicated negative torque, and the torques on numbers 3 and4 were 3,500 and 7,500 in-lbs, respectively. Moments earlier, the torques for numbers 3and 4 had been even lower. At 18:45:12, the pilot initiated a left hand turn toward theCalifornia coast.

6.3.3.3 Clue Three: Shortly after initiating the turn, the torques on engines 3 and 4began to recover. By 18:46:00, the torques on numbers 3 and 4 recovered briefly to 10,000and 11,000 in-lbs respectively. Flight test data show that this surging recovery isconsistent with fuel being intermittently supplied to the engines. Shortly after King 56completed the turn, torques on number 3 and 4 engines fell again, but this time rapidly.

6.3.3.4 Clue Four: Torque on engine number four recovered again, albeit briefly, forthe last time at 18:46:23. This torque recovery followed a brief dip of the right handwing tip. Once again, flight test data show that this surging recovery is consistent withfuel being intermittently supplied to the engines.

6.3.3.5 Clue Five: Despite the fact that torque fluctuated widely, all engine RPMindications remained normal throughout the whole sequence. This is a result of properpropeller blade angle governing. Individual RPMs only began to drop when their respectiveengines produced insufficient torque to maintain the engine and propeller at 100% RPM.

6.3.4 Initial Fuel Configuration: Based on the Form F and fuel servicing documentation,the BAR believes the mishap aircraft was serviced to approximately the following fuelload:

Table 6-2

King 56 Probable Fuel Loading

No. 1 Main No. 2 Main Lft Aux Rt Aux No. 3 Main No. 4 Main

7,000 lbs 7,000 lbs 4,000 lbs 0 lbs 7,000 lbs 7,000 lbs

Lft External Rt External

0 lbs 0 lbs

Left Fuselage Right Fuselage

0 lbs 8,000 lbs

6.3.5 The BAR believes the left hand external tank was not used because the fuelquantity indicator drove off scale, low end, when its forward boost pump was turned on.Fuel was not placed in the right hand external tank in order to maintain lateral balance.The right hand auxiliary tank was not used because it leaked around the cavity drain. Andfinally, the left hand fuselage tank was not used because the tank’s fuel quantitygauge, not the one on the flight deck, rotated backwards.

6.3.6 Regardless of the source of the fuel on King 56, calculations show that the totalamount of fuel required to fly the mishap profile is approximately 8,400 lbs. Thesecalculations were performed using T.O. 1-C-130H-1-1, and the specific amount of fuelrequired for each phase of the flight for the mishap sortie is shown in Table 6-3 below:

Table 6-3

Calculated Fuel Requirements for King 56

Time

(minutes)

Burn Rate

(1,000 lbs/hr)

Fuel Used

(1,000 lbs)

Fuel

Remaining

Start, Taxi

15

3.00

0.75

39.25

Takeoff

2

Tables

0.30

38.95

Climb

20

Tables

2.35

36.60

Cruise

64

4.68

5.00

31.60

6.3.7 The validity and accuracy of these calculations was investigated further in aC-130 simulator where all the specifics for the mishap sortie (i.e., aircraft grossweight, aircraft drag, weather conditions, etc.) were input into the simulator. Thesimulator flight crew was then tasked to fly the mishap sortie profile. This includedstarting engines, taxi to the runway, taking off, climbing, leveling off and cruising, andperforming all required checklists. Based upon DFDR data from King 56, the total time toperform engine start, taxi and takeoff in the simulator was nearly identical to that atPortland. Simulator time-to-climb was slightly faster than that on the DFDR--18:10 and18:00 minutes in the simulator versus 20:00 minutes for King 56. Cruise performance (i.e.,indicated airspeed and engine torques) was eventually "matched" with appropriatepositioning of the throttles.

6.3.8 Fuel requirements from the simulator were very similar to those shown in Table6-3. Fuel required for start, taxi and takeoff in the simulator was nearly 800 lbs versus1,050 pounds calculated. Fuel required for climb in the simulator was nearly 2,100 lbsversus 2,350 calculated. Fuel flow readings taken during the cruise phase in the simulatoraveraged at 4,700 lbs/hr for all engines versus 4,680 lbs/hr calculated.

6.3.9 The 250 lb differences between simulator results and calculations for both thestart, taxi, takeoff and climb phases are not considered significant given the certaintyof the data used (i.e., aircraft gross weight, aircraft drag, weather conditions, engineperformance, etc.) and the fidelity and layout of the performance charts themselves. Inshort, the simulator verified, within a given tolerance, that the calculated amount offuel required to fly the mishap profile presented in Table 6-3 is correct.

6.3.10 Fuel System: An extensive fuel system positive fuel flow check is performed bythe flight engineer prior to the first sortie of the day and after the aircraft isrefueled to ensure proper engine operation. In addition, two pressure gauges, four lowfuel pressure warning lights, and several tank empty caution lights are exercised duringthis check. If an aerial refueling off-load is planned, an aerial refueling manifoldpressurization check must also be performed.

6.3.11 Fuel system operation is categorized in one of two ways. Fuel sent directly fromthe main tanks to their respective engines is called "tank-to-engine" operation.In this case, the number one main tank feeds only the number one engine, and so on for theother three engines. During tank-to-engine operation, fuel from other tanks (i.e.,external, auxiliary and fuselage) is not used. The alternate method of fuel systemoperation is called "crossfeed." In this situation, any of the other"non-main" tanks can feed any of the engines. It is also possible to crossfeedfrom the main tanks as well. All crossfeed operations involve the use of the crossfeedmanifold. Crossfeeding from the fuselage tanks also involves the use of a fuselage tankmanifold, the refuel/dump manifold, and the right hand external crossfeed manifold. Priorto crossfeeding operations, the manifolds must be primed--that is purged of all air.Priming procedures are contained in T.O. 1C-130(H)H-1, Section 7. This procedure basicallyentails turning on the source pump(s), holding the crossfeed prime button depressed whichopens a valve to the number two main tank, and the crossfeed separation valve, andallowing sufficient time for the pump(s) to fill the manifold while exhausting anyair/fuel mixture into the number 2 main tank.

6.3.12 The fuel pumps in the auxiliary, external and fuselage tanks remain turned offunless they are being used. This is not true for the main tanks. For the main tanks, thepumps remain on whether the main tanks are being used or not. Exceptions to this"rule" are made for handling fuel imbalances (i.e., one main tank pump off) andemergencies (i.e., two main tank pumps off). Under normal operating conditions, there isno reason to intentionally turn off three or four main tank pumps at the same time inflight. Always operating with the main tank pumps on ensures that the engines will receivean uninterrupted pressurized supply of fuel, should an unplanned event occur duringcrossfeed operations, such as: an auxiliary, external, or fuselage tank running dry; anauxiliary, external, or fuselage tank pump fails; or, the crossfeed or refuel/dumpmanifolds fail or are somehow blocked.

6.3.13 Main tank pump output pressure is 15-24 psi. Auxiliary, external and fuselagetank pump output pressure is 28-40 psi. By virtue of the differences, when the auxiliary,external or fuselage tanks are selected on crossfeed, they will feed the engines eventhough the main tank pumps are on. Their higher pressure overrides the output pressurefrom the main tanks, thereby preventing fuel flow from the main tanks.

6.3.14 Low fuel pressure warning lights illuminate when aircraft fuel pressure to theirrespective engine falls below 8.5 psi. Tank empty caution lights for the auxiliary,external, and fuselage tanks come on when pump output pressure for their respective tankfalls below 23 psi. The main tanks do not have tank empty lights since they should alwayscontain some amount of fuel.

6.3.15 Only main tanks have the ability to gravity feed. Gravity feed operation ischecked in flight during functional check flights (FCF). During the gravity feed check,the aircraft is in tank-to-engine and fuel flow to all engines is established. Then thenumber one main tank pump is turned off. Proper engine operation under gravity feed isconfirmed, and the number one main tank pump is turned back on. This process is repeatedfor the remaining main tanks in sequence and takes approximately one minute for each. TheHC-130 aircraft also check gravity feed as part of the positive fuel flow check prior toflight.

6.3.16 In this mishap, there are essentially two basic sequences in which the flightengineer might have planned to use fuel. One sequence uses the mains first (for taxi,takeoff, and some portion of the climb), followed by the left hand auxiliary, right handfuselage, and then back to the main tanks again. Alternatively, he could have used themains, followed by the right hand fuselage, left hand auxiliary, and then back to the maintanks. Conventional wisdom says that the former method would most likely have been chosen.This is due to the fact that auxiliary tanks have only one pump and there is nopossibility, including gravity feed, to obtain fuel from them in the event of a pumpfailure. Since the planned mission was over water, the crew would want to burn auxiliarytank fuel early or know as soon as possible if that fuel would be unusable (i.e., trapped)during the mission.

6.4 Scenarios Evaluated

6.4.1 The BAR identified 20 possible scenarios that could have occurred on 22 Nov 96resulting in the loss of HC-130P, 64-14856 and all but one of the people aboard. Each ofthese possible scenarios is presented below and includes: (1) details of the scenario or abrief synopsis, (2) a numbered step-by-step failure sequence including detailedinformation for each step, when appropriate, and (3) corroborating and rebutting evidence,data and rationale. For brevity, the numbered step-by-step failure sequences for manyscenarios do not include detailed information because this information was alreadypresented in a previous scenario. The same is true for repeated corroborating andrebutting evidence, data and rationale. Additionally, the reader is referred to summarywrite-ups and more detailed documents when appropriate. This same format is also used foradditional physical evidence needed to further corroborate or rebut each scenario.

6.4.1.1 Scenario Number 1. Left Hand Auxiliary Fuel Tank Run Empty

6.4.1.1.1 Scenario Details: With the preceding fuel load in Table 6-2, it ispresumed that the engines fed from their respective main tanks for approximately the first33 minutes of flight. For roughly the next 51 minutes, it is postulated that the engineswere fed fuel from the left hand auxiliary tank. At this point, the aircraft is one hourand 24 minutes into the flight--the time when the flight engineer needs to use fuel fromthe right hand fuselage tank and the same time when unexplained torque fluctuations occur.See Table 6-4 for fuel-related details about each phase of the sortie for this scenario.

Time

(min)

Fuel Used

(k lbs)

Fuel Source, Duration and Quantity Used
Start, Taxi

15

0.75

Tank-to-Engine, 15 min, 750 lbs
Takeoff

2

0.30

Tank-to-Engine, 2 min 300 lbs
Climb

20

2.35

Tank-to-Engine, 20 min, 2350 lbs
Cruise

64

5.00

Tank-to-Engine, 13 min, 1000 lbs then Left Hand Auxiliary, 51 min, 4000 lbs

Table 6-4: Fuel Burn Profile for Scenario Number 1

Fuel flow to the engines may have ceased in the following manner:

6.4.1.1.1.1 All the main tank pumps are off. Turning all the main tank pumps off is notan acceptable technique because positive fuel flow to the engines is not ensured in theevent certain fuel system events, like those discussed earlier, occur. This technique isnot taught to flight engineers during formal training, but the BAR identified severalinstances where its use was confirmed or strongly suspected. Additionally, the main tankpumps could have been inadvertently left off. The BAR discovered instances where this hasoccurred. Whether the pumps were off intentionally or inadvertently is not important. Thekey point is that the switches were in the off position. Lastly, all four main tank pumpscould have failed or lost their power sources (each of the four main tank pumps has adifferent AC electrical power source) during the mishap sortie resulting in the pumps notoperating. Although theoretically possible, this did not happen to King 56 because theDFDR and the CVR were continually powered until the last generator fell off line..

6.4.1.1.1.2 All the engines are being crossfed from the left hand auxiliary tank and itis pumped empty. Normally, flight engineers transition from tank-to-engine to crossfeedshortly after takeoff. Given that King 56 contained 40,000 lbs of fuel and only 20,000 lbswas planned for use on the sortie to Naval Air Station North Island, there may have beenno urgency to transition to crossfeed operations. Additionally, the perceived presence ofcrossing air traffic, identified by Seattle Control during the climb to FL220, may havedelayed the transition to crossfeed operations if the flight engineer helped scan outsidethe cockpit. For these reasons it is presumed that tank-to-engine was utilized for thefirst 33 minutes of flight before crossfeeding from the left hand auxiliary tank for theremaining 51 minutes.

6.4.1.1.1.3 Illuminated fuel warning/caution lights are either not seen or notbelieved. If the left hand auxiliary tank was pumped empty, five warning/caution lights onthe overhead fuel panel should illuminate nearly simultaneously. They are the four lowfuel pressure warning lights as well as the left hand auxiliary tank empty caution light.Although the fuel panel is not in an ideal location, these warning/caution lights shouldhave been visible to the aircrew. However, they may not have been seen or acted upon forseveral reasons. First, the crew may have been too focused on the engine instruments anddid not see these illuminated overhead warning/ caution lights. Second, since this was anight over water navigational training/evaluation mission, it is possible the warning andcaution light switch may have been in the DIM position making the lights less apparent.Third, the survivor testified that the flight engineer had been reading a book prior tothe problem surfacing. It is possible his eyes may have been slow to adjust to the darkercockpit surroundings thereby increasing the time required before he could see the overheadwarning/caution lights. If the warning and caution light switch was in DIM, this situationwould only be further aggravated. Lastly, even if the five fuel warning/caution lightswere seen, it is possible that they were not recognized as the early warning signs of animpending fuel starvation. Instead, it is possible they may have been viewed asindications resulting from either a bleed air or an electrical problem. We know from thecockpit voice recorder that the pilot mentioned that they were perhaps dealing with ableed air or an AC problem.

6.4.1.1.1.4 Gravity feed does not establish itself from the main tanks. When the lefthand auxiliary tank (or any crossfed tank for that matter) runs empty, it is commonlybelieved that main tanks will automatically gravity feed to the engines. Despite the factthat the ability to gravity feed is checked, this check does not indicate that gravityfeed can be established from the main tanks when previous fuel flow to the engines wasthrough the crossfeed manifold. During the gravity fuel flow check, flow to a particularengine is first established with its associated main tank pump and then that pump isturned off. Fuel flow continues because a direct flow path from the tank to the engine hasbeen established and this flow has momentum. This check in no way involves the crossfeedmanifold. With a flow path established in the crossfeed manifold, some of thetank-to-engine flow path contains stagnant fuel which has no momentum. In this case, it isnot only necessary to establish a new flow path (i.e., tank-to-engine), but its velocitymust be increased sufficiently so that engine fuel flow demands can be met. If they arenot met, the engine will initially lose power. If this condition persists long enough, theengine may flameout. There is no data to show that gravity feed can be established at allpoints in the aircraft’s operating envelope given that the main tank pumps are offand flow from the crossfeed manifold ceases.

6.4.1.1.1.5 Air enters the fuel supply manifold and is delivered to all four engines.Air in a fuel supply line has the potential to adversely affect critical engine fuelcontrol components, disturb/disrupt combustion within the engines, or cavitate engineboost pumps. A small amount of air can be readily consumed by an engine and no one isaware this has occurred provided there is no loss in power. If the amount of air weregreater, it may adversely affect the fuel control, adversely affect the temperature datumsystem, or disturb combustion within the engine temporarily causing the engine to losepower. If the amount of air were greater yet, it may disrupt combustion within the enginelong enough to result in a flameout. However, at some point a large enough amount of airwill cavitate an engine driven fuel pump, disrupting fuel flow and causing the engine toflameout. The precise failure mechanism, due to an undetermined amount of air introducedinto the fuel supply manifold, is not known.

6.4.1.1.1.5.1 If all the main tank pumps are off and the left hand auxiliary tank isrun empty, it is possible that either an adequate gravity feed will not establish itselfat 22,000 feet or air will enter the crossfeed manifold resulting in multiple engineflameouts. T.O. 1C-130(H)H-1, page 3-24 states, "If a partial tank and an empty tankare on crossfeed with the boost pump inoperative in the partial tank, the engine being fedfrom the empty tank will be starved by air being drawn into the fuel line." Or statedmore specifically in terms of this scenario, if the main tanks and an empty auxiliary tankare on crossfeed with the boost pumps failed or off in the main tanks, the engine beingfed from the empty auxiliary tank will be starved by air being drawn into the fuel line.The BAR has been unable to uncover any existing test reports or analysis thatsubstantiates this sentence in T.O. 1C-130(H)H-1.

6.4.1.1.1.6 All four engines eventually flameout. If no corrective actions were takento reestablish fuel flow to the engines, all four engines would flameout. The primarycorrective action necessary is the immediate restoration of positive fuel flow to theengines. This could be accomplished by turning the main tank fuel boost pumps on andreturning to tank-to-engine.

6.4.1.1.2 Evidence, Data and Rationale: This scenario is corroborated by thecrew’s recorded comments pertaining to fuel flow, the fact that engines were lostsequentially not simultaneously, the incidents believed to be caused by fuel starvationthat were detailed in Section 5, the incident history and continuing possibility that maintank pumps can be inadvertently left off, and the results of ground testing performed atthe request of the BAR (i.e., Allison’s air injection testing at Little Rock AFB andAir Force testing at Edwards AFB). See Section 7 or the specific test report document ofinterest for more details on the results of individual tests.

6.4.1.1.2.1 The flight test performed at Edwards AFB rebuts this scenario. Althoughengines being crossfed from a left hand auxiliary tank when it is run empty experiencedtorque fluctuations on the ground, torque fluctuations were not observed in flight.Additionally, analysis of radar information reveals that there was no traffic close toKing 56 during its climb. This makes it less likely that the flight engineer was scanningoutside the cockpit thereby delaying the transition from tank-to-engine to crossfeedoperations. The flight engineer had sufficient time during the two minute hold down at15,000 feet to transition from tank-to-engine to crossfeed operations. Lastly, testimonyfrom flight engineers who knew the mishap flight engineer had never heard him discuss atechnique where all the main tank pumps would be turned off.

6.4.1.1.2.2 Physical evidence needed to further corroborate or rebut this scenario are:

1. The Wing Section. From the recovery videos, this section of the wing contains the crossfeed separation valve, the left hand auxiliary tank crossfeed valve and the number 2 main tank crossfeed valve. It probably also contains the number 1 main tank crossfeed valve. These are four of the six valves that would normally be open if the left hand auxiliary tank was crossfeeding to all four engines. The other two valves (the number 3 and number 4 main tank crossfeed valves) are no longer attached to this piece of wreckage. All of these valves are DC powered, and upon the loss of the last engine generator, their positions are captured. Since the cockpit voice recorder is devoid of any discussion of engineer overhead fuel control panel actions, the BAR believes that the valves are in the same positions as when the power loss first started. The position of these valves will help determine the source of fuel to the engines upon the loss of the last engine generator.

2. The Forward Fuel Control Panel and Auxiliary Fuel Panel. These panels contain numerous, lights, switches, control knobs and fuel quantity gauges. The lights, switches and control knobs are of little interest because they could have, and probably were, moved several times in an effort to restart the engines during the unpowered descent to the ocean. However, the fuel quantity gauges do not respond to manipulation of these switches and knobs. They only respond to fuel quantity within the tanks. These gauges are AC powered and, upon the loss of last engine generator, retain their last reading. Therefore, the final fuel state on the aircraft might be determined from these gauges.

3. Both Fuselage Tanks. These tanks each contain a control panel used for ground refueling. This control panel also contains a tank fuel quantity gauge. The gauges from both of these tanks are desired for the same reasons the cockpit fuel gauges are desired. Although one of these tanks was initially fueled and the other empty, it is difficult to distinguish between the two tanks making recovery of both tanks desirable.

6.4.1.2 Scenario Number 2. Right Hand Fuselage Fuel Tank Run Empty

6.4.1.2.1 Scenario Details (Variation A): This scenario is a variation ofscenario number 1. In this case, the supposition is that the right hand fuselage tank isrun empty instead of the left hand auxiliary tank. This could happen in one of two ways.First, with the preceding fuel load in Table 6-2, it is presumed that the engines fed fromtheir respective main tanks for taxi, takeoff, and approximately 10 minutes into climbbefore transitioning to crossfeed operations. In this case, it is postulated that theengines were crossfed fuel from the right hand fuselage tank for the remainder of thesortie. To fly the remaining 10 minutes of climb and 64 minutes of cruise at FL220, it isestimated that 6200 lbs of fuel is required, 1800 lbs less than that believed to be in theright hand fuselage tank. See Table 6-5 for fuel-related details about each phase of thesortie for this scenario. With fuel quantity gauging/indicating failures, it is possiblethat 8000 lbs indicated really equated to 6200 lbs actual and the right hand fuselage tankis now empty. It is also possible that this 1800 lbs was inadvertently transferred toother tanks, via broken or leaking valves, rather than being burned. Reference T.O.1C-130(H)H-1, page 3-23 Warning and page 7-7 Note which states, "Any time therefuel/dump manifold is pressurized for any reason, fuel can transfer into any tank. Thisis possible due to leakage through a check valve, a malfunctioning refuel shutoff floatvalve, or loose connections on the fuel line in the tank. When this manifold ispressurized, all fuel quantity indicators must be closely monitored for an unusual changein quantity, or if fuel indicators are inoperative, fuel flow versus fuel quantity andlateral trim must be monitored." At this point, the aircraft is one hour and 24minutes into the flight--a time when the flight engineer needs to use fuel from the lefthand auxiliary tank and the same time when unexplained torque fluctuations occur. If themain tank pumps were turned off and the fuel panel warning/caution lights not seen orbelieved, then the engines could have flamed out due to pressurized cabin air entering thefuel supply manifold via either the empty left hand or right hand fuselage tank, whichworked its way out to all four engines.

Time

(minutes)

Fuel Used

(k lbs)

Fuel Source, Duration and Quantity Used
Start, Taxi

15

0.75

Tank-to-Engine, 15 min, 750 lbs
Takeoff

2

0.30

Tank-to-Engine, 2 min 300 lbs
Climb

20

2.35

Tank-to-Engine, 10 min, ~1150 lbs then Right Hand Fuselage, 10 min, ~1200 lbs
Cruise

64

5.00

Right Hand Fuselage, 64 min, 5000 lbs

Table 6-5. Fuel Burn Profile for Scenario Number 2.a

Fuel flow to the engines may have ceased in the following manner:

1. The aircraft either has right hand fuselage tank fuel quantity gauging/indicating problems or leaking/failed fuel valves.

2. All the main tank pumps are off.

3. All the engines are being crossfed from the right hand fuselage tank and it is pumped empty.

4. Illuminated fuel warning/caution lights are either not seen or not believed.

5. Gravity feed does not establish itself from the main tanks.

6. Cabin air enters the fuel supply manifold and is delivered to all four engines.

7. All four engines eventually flameout.

6.4.1.2.2 Scenario Details (Variation B): The second way to run the right handfuselage tank empty may have occurred as follows. For an unknown reason, the entiresortie, including taxi, takeoff, climb and cruise, were flown while crossfeeding all fourengines from the right hand fuselage tank. To fly this profile requires approximately 8400lbs of fuel, 400 lbs more than that believed to be in the right hand fuselage tank. Withfuel quantity gauging/indicating tolerances, it is possible that 8000 lbs indicated reallyequated to 8400 lbs actual and the right hand fuselage tank is now empty. At this point,the aircraft is one hour and 24 minutes into the flight--the time when the flight engineerneeds to use fuel from the left hand auxiliary tank and the same time when unexplainedtorque fluctuations occur. If the main tank pumps were turned off and the fuel panelwarning/caution lights not seen or believed, then the engines could have flamed out due topressurized cabin air entering the fuel supply manifold via either the empty left hand orright hand fuselage tank, which worked its way out to all four engines. Fuel flow to theengines may have ceased in the following manner:

1. All the main tank pumps are off.

2. All the engines are being crossfed from the right hand fuselage tank and it is pumped empty.

3. Illuminated fuel warning/caution lights are either not seen or not believed.

4. Gravity feed does not establish itself from the main tanks.

5. Cabin air enters the fuel supply manifold and is delivered to all four engines.

6. All four engines eventually flameout.

6.4.1.2.3 Evidence, Data and Rationale: This scenario (both variations) iscorroborated by the crew’s recorded comments pertaining to fuel flow, the fact thatengines were lost sequentially not simultaneously, the aircraft’s recent history offuel gauging problems with some tanks that were being carried as open write-ups, theincidents believed to be caused by fuel starvation that were detailed in Section 5, theincident history and continuing possibility that main tank pumps can be inadvertently leftoff, the results of ground testing performed at the request of the BAR (i.e.,Allison’s air injection testing at Little Rock AFB), and the results of flighttesting performed at the request of the BAR (i.e., Air Force testing at Edwards AFB). SeeSection 7 or the specific test report document of interest for more details on the resultsof individual tests.

6.4.1.2.4 This scenario (both variations) is rebutted by testimony from flightengineers who knew the mishap flight engineer and had never heard him discuss a techniquewhere all the main tank pumps would be turned off. Physical evidence needed to furthercorroborate or rebut this scenario are:

1. The Wing Section. From the recovery videos, this section of the wing contains the crossfeed separation valve and the number 2 main tank crossfeed valve. It probably also contains the number 1 main tank crossfeed valve. These are three of the seven valves that would normally be open if the right hand fuselage tank was crossfeeding to all four engines. Also, the left hand auxiliary tank crossfeed valve, which is part of the wing, should be found closed. The other four valves, the number 3 and number 4 main tank crossfeed valves, the right hand external dump valve and the right hand external tank crossfeed valve are no longer attached to this piece of wreckage. All of these valves are DC powered, and upon the loss of the last engine generator, their positions are captured. Since the cockpit voice recorder is devoid of any discussion of fuel panel actions, the BAR believes that the captured valve positions are the same valve positions as when the power loss first started. The position of these valves will help determine the source of fuel to the engines upon the loss of the last engine generator.

2. The Forward Fuel Control Panel and the Auxiliary Fuel Panel.

3. Both Fuselage Tanks.

6.4.1.3 Scenario Number 3. Insufficient Fuel Manifold Priming

6.4.1.3.1 Scenario Details: The fuel burn for this scenario is assumed to be thesame as previously described in scenario number 1. The engines fed from their respectivemain tanks for approximately the first 33 minutes of flight. This is followed by 51minutes of all four engines being fed fuel from the left hand auxiliary tank. At thispoint, the aircraft is one hour and 24 minutes into the flight--the time when the flightengineer needs to use fuel from the right hand fuselage tank and the same time whenunexplained torque fluctuations occur. See Table 6-4 for fuel-related details about eachphase of the sortie for this scenario. A switch from the left hand auxiliary tank to theright hand fuselage tank necessitates that the crossfeed manifold, the right external tankmanifold, the refuel/dump manifold, and the fuselage tank manifold be primed, or purged ofair. According to T.O. 1C-130(H)H-1, page 7-7, priming is required for 30 seconds. Sincethe crossfeed manifold was successfully used while crossfeeding from the left handauxiliary tank, it is unlikely it contains any air. However, to get fuel from the righthand fuselage tank also requires the use of three other manifolds. If any of these othermanifolds contained significant amounts of air and was not primed sufficiently, remainingair may cause the engines to flameout. Specifically, if the right hand fuselage tank had afailed fuel level control valve that leaked, or two check valves that leaked (all of theseitems are contained within the fuselage tank), then it is possible to drain, via gravity,much of the fuel in the refuel/dump and fuselage tank manifolds back into the right handfuselage tank. (This same potential failure mode also exists for the left hand fuselagetank.) As fuel drains from these manifolds, air is allowed to enter into them which mustbe eliminated via priming. As a final point, it is also worthy to note that with theaddition of fuselage tanks to the C-130, it appears that there was no correspondingincrease in the time required to prime the longer fuel supply manifold. In short, 30seconds of priming may be an insufficient amount of priming time. Fuel flow to the enginesmay have ceased in the following manner:

1. Finished crossfeeding from left hand auxiliary tank.

2. Preparing to use right hand fuselage tank.

3. Manifold not primed sufficiently--air remaining.

4. Significant amount of air routed to all four engines.

5. All four engines eventually flameout.

6.4.1.3.2 Evidence, Data and Rationale: This scenario is corroborated by thecrew’s recorded comments pertaining to fuel flow, the fact that engines were lostsequentially not simultaneously, the results of ground testing performed at the request ofthe BAR (i.e., Allison’s air injection testing at Little Rock AFB), and the resultsof flight testing performed at the request of the BAR (i.e., Air Force testing at EdwardsAFB). The BAR also spoke to several flight engineers who have experienced torquefluctuations as a result of inadequate priming, including one who had flamed out an engineas a result. See Section 7 or the specific test report document of interest for moredetails on the results of individual tests.

6.4.1.3.3 This scenario is rebutted by flight testing performed at the request of theBAR (i.e., Air Force testing at Edwards AFB, CA), and by the belief that the flightengineer would surely have made the connection with the transition to a new tank and theengine power loss now occurring. With the connection recognized, it is believed the flightengineer would quickly undo that just done and sought another source of fuel.Additionally, we know from the survivor’s testimony that the flight engineer had beenreading a book in the minutes before the power loss occurred. Furthermore, from thecockpit voice recorder, we know the flight engineer was engaged in a discussion with othercrew members at the time when the power loss started. All of these factors make itunlikely that improper manifold priming resulted in this mishap. Physical evidence neededto further corroborate or rebut this scenario are:

1. The Wing Section.

2. The Forward Fuel Control Panel and the Auxiliary Fuel Panel.

3. Both Fuselage Tanks.

6.4.1.4 Scenario Number 4. Right Hand Fuselage Fuel Tank Pump(s) Failure

6.4.1.4.1 Scenario Details: With the preceding fuel load in Table 6-2, it ispresumed that the engines fed from their respective main tanks for approximately the first10 minutes of flight. For the remaining 10 minutes of climb and roughly the next 36minutes of cruise, it is postulated that the engines were fed fuel from the left handauxiliary tank. Once the left hand auxiliary tank was emptied, it is presumed that asuccessful transition to the right hand fuselage tank was accomplished. At one hour and 24minutes into the flight, it is presumed that the right hand fuselage fuel tank pump(s)failed--the same time when the unexplained torque fluctuations occur. See Table 6-6 forfuel-related details about each phase of the sortie for this scenario. It is also worthyto note that the left hand auxiliary tank may not have been used at all in favor of theright hand fuselage tank during climb and cruise. In this case the right hand fuselagetank would have burned 6200 lbs of fuel when the pump(s) failed.

Time

(minutes)

Fuel Used

(k lbs)

Fuel Source, Duration and Quantity Used
Start, Taxi

15

0.75

Tank-to-Engine, 15 min, 750 lbs
Takeoff

2

0.30

Tank-to-Engine, 2 min 300 lbs
Climb

20

2.35

Tank-to-Engine, 10 min, ~1150 lbs then Left Hand Auxiliary, 10 min, ~1200 lbs
Cruise

64

5.00

Left Hand Aux, 36 min, ~2800 lbs then Right Hand Fuselage, 28 min, ~2200 lbs

Table 6-6. Fuel Burn Profile for Scenario Number 4

6.4.1.4.1.1 This scenario presumes that all four engines are being crossfed from theright hand fuselage tank when the pump(s) in that tank fails. Most HC-130P fuselage tankscontain two fuel pumps which can be controlled separately. If only one pump was turned on,which is normal practice, then only one pump must fail to support this scenario. If twopumps were turned on, then both pumps must fail. If the main tank pumps were turned offand the fuel panel warning/caution lights not seen or believed, then the engines couldhave flamed out due to pressurized cabin air, entering the fuel supply manifold via theempty left hand fuselage tank, which worked its way out to all four engines. Fuel flow tothe engines may have ceased in the following manner:

1. All the main tank pumps are off.

2. All the engines are being crossfed from the right hand fuselage tank.

3. The right hand fuselage tank pump(s) fails.

4. Illuminated fuel warning/caution lights are either not seen or not believed.

5. Gravity feed does not establish itself from the main tanks

6. Cabin air enters the fuel supply manifold and is delivered to all four engines.

7. All four engines eventually flameout.

6.4.1.4.2 Evidence, Data and Rationale: This scenario is corroborated by thecrew’s recorded comments pertaining to fuel flow, the fact that engines were lostsequentially not simultaneously, the incidents believed to be caused by fuel starvationthat were detailed in Section V, the incident history and continuing possibility that maintank pumps can be inadvertently left off, the results of ground testing performed at therequest of the BAR (i.e., Allison’s air injection testing at Little Rock AFB), andthe results of flight testing performed at the request of the BAR (i.e., Air Force testingat Edwards AFB). See Section 7 or the specific test report document of interest for moredetails on the results of individual tests.

6.4.1.4.3 This scenario is rebutted by testimony from flight engineers who knew themishap flight engineer and had never heard him discuss a technique where all the main tankpumps would be turned off. Physical evidence needed to further corroborate or rebut thisscenario are:

1. The Wing Section.

2. The Forward Fuel Control Panel and the Auxiliary Fuel Panel.

3. Both Fuselage Tanks. In addition to the items previously addressed for fuselage tanks, the fuel pumps from the right hand fuselage tank are also of interest. It is anticipated that both of these pumps would be examined to help determine if either had failed while operating.

6.4.1.5 Scenario Number 5. Left Hand Auxiliary Fuel Tank Pump Failure

6.4.1.5.1 Scenario Details: This scenario is a variation of scenario number 4.This scenario presumes that all four engines are being crossfed from the left handauxiliary tank when the pump in that tank fails. To get to this point, a fuel burn similarto that of scenario number 1 could have been used except that the transition to left handauxiliary tank would have been delayed beyond the 13 minutes after level off cited forscenario number 1. Although the transition could have been accomplished at any time beyond13 minutes after level off, we have assumed that the transition was made 25 minutes afterlevel off for the sake of discussion. See Table 6-7 for fuel-related details about eachphase of the sortie for this scenario. If the main tank pumps were turned off and the fuelpanel warning/caution lights not seen or believed, then the engines could have beenstarved of fuel due to an insufficient fuel flow.

Time

(minutes)

Fuel Used

(k lbs)

Fuel Source, Duration and Quantity Used
Start, Taxi

15

0.75

Tank-to-Engine, 15 min, 750 lbs
Takeoff

2

0.30

Tank-to-Engine, 2 min 300 lbs
Climb

20

2.35

Tank-to-Engine, 20 min, 2350 lbs
Cruise

64

5.00

Tank-to-Engine, 25 min, 1950 lbs then Left Hand Auxiliary, 39 min, 3050 lbs

Table 6-7. Fuel Burn Profile for Scenario Number 5

Fuel flow to the engines may have ceased in the following manner:

1. All the main tank pumps are off.

2. All the engines are being crossfed from the left hand auxiliary tank.

3. The left hand auxiliary tank pump fails.

4. Illuminated fuel warning/caution lights are either not seen or not believed.

5. Gravity feed does not establish itself from the main tanks.

6. All four engines eventually flameout.

6.4.1.5.2 Evidence, Data and Rationale: This scenario is corroborated by thecrew’s recorded comments pertaining to fuel flow, the fact that engines were lostsequentially not simultaneously, the incidents believed to be caused by fuel starvationthat were detailed in Section 5, the incident history and continuing possibility that maintank pumps can be inadvertently left off.

6.4.1.5.3 The results of ground and flight testing performed at the request of the BAR(i.e., Air Force testing at Edwards AFB) rebuts this scenario. In both cases, when pumpfailures were simulated in the left hand auxiliary tank, gravity feed successfullyestablished itself. Additionally, testimony from flight engineers who knew the mishapflight engineer had never heard him discuss a technique where all the main tank pumpswould be turned off. See Section 7 or the specific test report document of interest formore details on the results of individual tests. Physical evidence needed to furthercorroborate or rebut this scenario are:

1. The Wing Section. In addition to the items previously addressed, the fuel pump from the left hand auxiliary tank is also of interest. It is anticipated that this pump would be examined to help determine if it failed while operating.

2. The Forward Fuel Control Panel and the Auxiliary Fuel Panel.

3. Both Fuselage Tanks.

6.4.1.6 Scenario Number 6. Undetected Fuel Leak

6.4.1.6.1 Scenario Details: This scenario assumes that an undetected fuel leakexisted in the refuel/dump manifold. In this scenario, the engines are fed normally fromtheir respective main tanks during takeoff and some portion of climb. The transition toand use of the left hand auxiliary tank is successful. The transition to the right handfuselage tank is also successful. (Note: The refuel/dump manifold is not pressurized whencrossfeeding from the left hand auxiliary tank but is pressurized when crossfeeding fromthe right hand fuselage tank.) However, at some point, a large fuel leak develops in therefuel/dump manifold outside of the wing (i.e., aft of the aft spar). Unbeknownst to thecrew, this leak quickly drains the right hand fuselage tank as well as all four maintanks. With no more fuel on board, all four engines flameout. This scenario assumes thatthe crew did not notice the problem early because the fuselage tank empty caution lightwas not seen or burned out. If the crew had noticed this light, they could have come offcrossfeed and gone to tank-to-engine without any further problems. Fuel flow to theengines may have ceased in the following manner:

1. Finished crossfeeding from left hand auxiliary tank.

2. Successfully transitioned to right hand fuselage tank.

3. Undetected leak in refuel/dump manifold aft of the aft spar.

4. Right hand fuselage tank emptied.

5. Fuselage tank empty caution light either not seen or burned out.

6. All four main tanks emptied.

7. All four engines eventually flameout.

6.4.1.6.2 Evidence, Data and Rationale: This scenario is corroborated by thecrew’s recorded comments pertaining to fuel flow, the fact that engines were lostsequentially not simultaneously, and there have been other incidents in the C-130 fleetwhere minor refuel/dump manifold leaks have gone undetected for a period of time.

6.4.1.6.3 This scenario is rebutted by US Coast Guard and US Navy observations that alarge petroleum slick was observed on the ocean’s surface during rescue and recoveryefforts. Additionally, US Coast Guard personnel reported that fuel was pouring from thewing during its attempted recovery. Physical evidence needed to further corroborate orrebut this scenario are:

1. The Wing Section.

2. The Forward Fuel Control Panel and the Auxiliary Fuel Panel.

3. Both Fuselage Tanks.

6.4.1.7 Scenario Number 7. Fuel Dump Valve(s) Stuck Open

6.4.1.7.1 Scenario Details: This scenario is a variation of scenario number 6and assumes that one or both of the fuel dump valves are stuck open. In this scenario, theengines are fed normally from their respective main tanks during takeoff and some portionof climb. The transition to and use of the left hand auxiliary tank is successful. Thetransition to the right hand fuselage tank is also successful. However, with therefuel/dump manifold now pressurized, fuel begins to be dumped overboard via one or bothdump masts near the wing tips. Unbeknownst to the crew, this uncommanded fuel dumpingquickly drains the right hand fuselage tank as well as all four main tanks. With no morefuel on board, all four engines then flameout. This scenario assumes that the crew did notnotice the problem early because the fuselage tank empty caution light was not seen orburned out. If the crew had noticed this light, they could have come off crossfeed andgone to tank-to-engine without any further problems. Fuel flow to the engines may haveceased in the following manner:

1. One or both of the fuel dump valves is stuck open.

2. Finished crossfeeding from left hand auxiliary tank.

3. Successfully transitioned to right hand fuselage tank.

4. Right hand fuselage tank emptied.

5. Fuselage tank empty caution light either not seen or burned out.

6. All four main tanks emptied.

7. All four engines eventually flameout.

6.4.1.7.2 Evidence, Data and Rationale: This scenario is corroborated by thecrew’s recorded comments pertaining to fuel flow, the fact that engines were lostsequentially not simultaneously, and the fact that there have been other incidents in theC-130 fleet where uncommanded dumping has gone undetected for a period of time.

6.4.1.7.3 This scenario is rebutted by US Coast Guard and US Navy observations that alarge petroleum slick was observed on the ocean’s surface during rescue and recoveryefforts. Additionally, US Coast Guard personnel reported that fuel was pouring from thewing during its attempted recovery. Physical evidence needed to further corroborate orrebut this scenario are:

1. The Wing Section.

2. The Forward Fuel Control Panel and the Auxiliary Fuel Panel.

3. Both Fuselage Tanks.

6.4.1.8 Scenario Number 8. Refuel/Dump Line Rupture

6.4.1.8.1 Scenario Details: This scenario is the same as scenario number 6except that the leak is assumed to be much larger. The leak is so large, that whencombined with a presumed low pressure area just behind the aft spar, fuel is actuallypulled/siphoned from the refuel/dump manifold to such an extent that an insufficient fuelflow is available to the engines and they flameout. This occurs even though there is stilla significant amount of fuel remaining in the right hand fuselage tank and main tanks.Fuel flow to the engines may have ceased in the following manner:

1. Finished crossfeeding from left hand auxiliary tank.

2. Successfully transitioned to right hand fuselage tank.

3. Undetected rupture of refuel/dump manifold aft of the aft spar.

4. Fuel is pulled/siphoned from the refuel/dump manifold.

5. Reduced fuel flow causes all four engines to eventually flameout.

6.4.1.8.2 Evidence, Data and Rationale: This scenario is corroborated by thecrew’s recorded comments pertaining to fuel flow and the fact that engines were lostsequentially not simultaneously.

6.4.1.8.3 Nothing currently rebuts this scenario. Physical evidence needed to furthercorroborate or rebut this scenario are:

1. The Wing Section.

2. The Forward Fuel Control Panel and the Auxiliary Fuel Panel.

3. Both Fuselage Tanks.

6.4.1.9 Scenario Number 9. Water in Right Hand Fuselage Fuel Tank

6.4.1.9.1 Scenario Details: Water can be inadvertently introduced into oraccumulate in fuselage tanks. The fuel burn for this scenario is assumed to be the same aspreviously described in scenario number 1. The engines fed from their respective maintanks for approximately the first 33 minutes of flight. This is followed by 51 minutes ofall four engines being fed fuel from the left hand auxiliary tank. At this point, theaircraft is one hour and 24 minutes into the flight--the time when the flight engineerneeds to use fuel from the right hand fuselage tank and the same time when unexplainedtorque fluctuations occur. See Table 6-4 for fuel-related details about each phase of thesortie for this scenario. Portland had experienced extremely rainy weather in the daysprior to the mishap. This rainfall may have worked its way into a fuel servicing vehiclethat was dispatched to service the King 56. Or equivalently, several temperature cyclesduring extremely humid conditions may have resulted in significant condensation inside thefuselage tanks. If this water was not drained from the right hand fuselage tank, it mayhave been sent to the engines resulting in four flameouts. Currently, there is norequirement to periodically sump fuselage tanks. Fuel flow to the engines may have ceasedin the following manner:

1. Water finds its way into the right hand fuselage tank.

2. Right hand fuselage fuel tank is not sumped prior to flight.

3. Finished crossfeeding from left hand auxiliary tank.

4. Successfully transitioned to right hand fuselage tank.

5. Water is sent to all four engines.

6. All four engines eventually flameout.

6.4.1.9.2 Evidence, Data and Rationale: This scenario is corroborated by thefact that engines were lost sequentially not simultaneously, the rainy weather at Portlandin the days prior to the mishap, and the fact that no requirement exists to sump (drainingat lowest point of the tank to check for and drain water and contaminants) fuselage tanks.

6.4.1.9.3 This scenario is rebutted by the performance of a positive fuel flow check(reference T.O. 1C-130(H)H-1, page 7-5) on the ground which would have identified water inthe tank, the crew’s recorded comments pertaining to fuel flow, and the fact thatlaboratory testing performed on fuel samples taken from the tank farm at Portland, thetruck that serviced King 56, the recovered number 4 engine, and recovered aircraft fueltank foam revealed nothing unusual. Additionally, other aircraft serviced by the sametruck used their fuel without incident. Physical evidence needed to further corroborate orrebut this scenario are:

1. The Wing Section.

2. The Forward Fuel Control Panel and the Auxiliary Fuel Panel.

3. Both Fuselage Tanks.

6.4.1.10 Scenario Number 10. Water in Left Hand Auxiliary Fuel Tank

6.4.1.10.1 Scenario Details: This scenario is a variation of scenario number 9.Water can be inadvertently introduced into or accumulate in auxiliary tanks. Portland hadexperienced extremely rainy weather in the days prior to the mishap. This rainfall mayhave worked its way into a fuel servicing vehicle that was dispatched to service King 56.Or equivalently, several temperature cycles during extremely humid conditions may haveresulted in significant condensation inside the auxiliary tank. If this water was notdrained from the left hand auxiliary tank, it may have been sent to all of the enginesresulting in four flameouts. T.O. 1C-130H-2-12JG-10-1, page 1-76 requires sumping of maintanks (only for those not modified by TCT.O. 1C-130-1309 and having a tail number below73-1580), auxiliary tanks and external tanks. The BAR talked to several flight linepersonnel at various bases who indicated that this does not routinely occur. Fuel flow tothe engines may have ceased in the following manner:

1. Water finds its way to the left hand auxiliary tank.

2. Left hand auxiliary fuel tank is not sumped prior to flight

3. Finished crossfeeding from right hand fuselage tank.

4. Successfully transitioned to left hand auxiliary tank.

5. Water is sent to all four engines.

6. All four engines eventually flameout.

6.4.1.10.2 Evidence, Data and Rationale: This scenario is corroborated by thefact that engines were lost sequentially not simultaneously, the rainy weather at Portlandin the days prior to the mishap, and testimony indicating that sumping does not routinelyoccur.

6.4.1.10.3 This scenario is rebutted by the performance of a positive fuel flow check(reference TO 1C-130(H)H-1, page 7-5) on the ground which would have identified water inthe tank, the crew’s recorded comments pertaining to fuel flow, and the fact thatlaboratory testing performed on fuel samples taken from the tank farm at Portland, thetruck that serviced King 56, the recovered number 4 engine, and recovered aircraft fueltank foam revealed nothing unusual. Additionally, other aircraft serviced by the sametruck used their fuel without incident. Physical evidence needed to further corroborate orrebut this scenario are:

1. The Wing Section.

2. The Forward Fuel Control Panel and the Auxiliary Fuel Panel.

3. Both Fuselage Tanks.

6.4.1.11 Scenario Number 11. Contaminated Fuel

6.4.1.11.1 Scenario Details: Periodically fuel becomes contaminated. Thiscontamination can come from the aircraft or an outside source. Regardless of the source,fine particulate contamination is of concern in this scenario. It is postulated that fineparticulates could work their way past the aircraft fuel filters, into the engine fuelcontrols, and over time, adversely affect the fuel metering valves inside the fuelcontrols. In certain circumstances, this could lead to engine flameouts. Fuel flow to theengines may have ceased in the following manner:

1. Fine particulate contamination finds its way into the engine fuel controls.

2. The contamination gradually accumulates in the fuel controls.

3. This contamination adversely affects the fuel metering valves.

4. All four engines eventually flameout.

6.4.1.11.2 Evidence, Data and Rationale: This scenario is corroborated by thecrew’s recorded comments pertaining to fuel flow, the fact that engines were lostsequentially not simultaneously, and there has been at least one other incident in theC-130 fleet where multiple engines have flamed out due to fuel contamination.

6.4.1.11.3 This scenario is rebutted by the fact that laboratory testing performed onfuel samples taken from the tank farm at Portland, the truck that serviced King 56, therecovered number 4 engine, and recovered aircraft fuel tank foam revealed nothing unusual.Physical evidence needed to further corroborate or rebut this scenario are:

1. Additional Engines. The recovery of additional engines may yield more clues about the possible presence of contaminated fuel. The low pressure filter, fuel pumps, fuel control, and temperature datum valve recovered with the number 4 engine are scheduled to be examined in January 1998.

6.4.1.12 Scenario Number 12. Right Hand Fuselage Fuel Tank Manual Isolation ValveClosed

6.4.1.12.1 Scenario Details: The fuel burn for thisscenario is assumed to be the same as previously described in scenario number 1. Theengines fed from their respective main tanks for approximately the first 33 minutes offlight. This is followed by 51 minutes of all four engines being fed fuel from the lefthand auxiliary tank. At this point, the aircraft is one hour and 24 minutes into theflight--the time when the flight engineer needs to use fuel from the right hand fuselagetank and the same time when unexplained torque fluctuations occur. See Table 6-4 forfuel-related details about each phase of the sortie for this scenario. The manifold thatconnects the fuselage tanks to the aircraft incorporates three manual isolation valves.One isolation valve simultaneously isolates both fuselage tanks from the aircraft fuelsystem. The other two valves only isolate one tank each. Since the left hand fuselage tankwas not fueled for the mishap sortie, the left hand manual isolation valve may have beenclosed thereby isolating it from the aircraft fuel system. However, assuming that theright hand fuselage tank was properly fueled, and that somehow its manual isolation valvewas closed either instead of or in addition to the left manual isolation valve, fuselagefuel would be trapped and not be available. Again, if the main tank pumps were off and thefuselage tank empty caution light burned out or not seen, then the engines could have beenstarved by an insufficient supply of fuel or pressurized cabin air may have enteredthrough the empty left hand fuselage tank. Fuel flow to the engines may have ceased in thefollowing manner:

1. The right hand manual isolation valve is inadvertently closed.

2. All the main tank pumps are off.

3. Finished crossfeeding from left hand auxiliary tank.

4. Unsuccessful transition to right hand fuselage tank.

5. Illuminated fuselage tank empty caution light either not seen or not believed.

6. Gravity feed does not establish itself from the main tanks.

7. Cabin air enters the fuel supply manifold and is delivered to all four engines.

8. All four engines eventually flameout.

6.4.1.12.2 Evidence, Data and Rationale: This scenario is corroborated by thecrew’s recorded comments pertaining to fuel flow, the fact that engines were lostsequentially not simultaneously, the incidents believed to be caused by fuel starvationthat were detailed in Section 5, the incident history and continuing possibility that maintank pumps can be inadvertently left off, and the results of ground testing performed atthe request of the BAR (i.e., Allison’s air injection testing at Little Rock AFB).Although not an exact replication of this scenario was tested at Edwards AFB, thepotential for pressurized cabin air to enter the fuel supply manifold and the potential toeventually flameout all four engines was demonstrated. See Section 7 or the specific testreport document of interest for more details on the results of individual tests.

6.4.1.12.3 This scenario is primarily rebutted by the procedure performed for the usageof fuselage fuel (reference T.O. 1C-130(H)H-1, page 7-7) which incorporates several stepsto ensure fuel is available, pressurized by a pump and then properly routed to theengines. If this procedure was performed improperly, it is believed that the flightengineer would surely have made the connection with the transition to a new tank and theengine power loss now occurring. With the connection recognized, it is believed the flightengineer would quickly undo that just done and sought another source of fuel.Additionally, we know from the survivor’s testimony that the flight engineer had beenreading a book in the minutes before the power loss occurred. Furthermore, from thecockpit voice recorder, we know that the flight engineer was engaged in a discussion withother crew members at the time when the power loss started. Moreover, fuselage tank manualisolation valves are normally left open even when the tanks are empty. Lastly, testimonyfrom flight engineers who knew the mishap flight engineer had never heard him discuss atechnique where all the main tank pumps would be turned off. All of these factors make itunlikely that a closed manual isolation valve resulted in this mishap. Physical evidenceneeded to further corroborate or rebut this scenario are:

1. The Wing Section.

2. The Forward Fuel Control Panel and the Auxiliary Fuel Panel.

3. Both Fuselage Tanks.

6.4.1.13 Scenario Number 13. Cabin Pressurization Procedure

6.4.1.13.1 Scenario Details: This procedure involves using cabin pressure, notfuselage tank pumps, to move fuel from the right hand fuselage tank to the engines. Themethodology to perform this procedure is covered in T.O. 1C-130(H)H-1, Section 7. The fuelburn for this scenario is assumed to be the same as previously described in scenarionumber 1. The engines fed from their respective main tanks for approximately the first 33minutes of flight. This is followed by 51 minutes of all four engines being fed fuel fromthe left hand auxiliary tank. At this point, the aircraft is one hour and 24 minutes intothe flight--the time when the flight engineer needs to use fuel from the right handfuselage tank and the same time when unexplained torque fluctuations occur. See Table 6-4for fuel-related details about each phase of the sortie for this scenario. Again, if allthe main tank pumps were off, not just a maximum of two allowed by the Section 7procedure, and the left hand fuselage tank manual isolation valve open, pressurized cabinair would be sent through the fuel supply manifold via the empty left hand fuselage tank.If allowed to persist long enough, this would result in four engine flameouts. Fuel flowto the engines may have ceased in the following manner:

1. All the main tank pumps are off.

2. Finished crossfeeding from left hand auxiliary tank.

3. Unsuccessful transition to right hand fuselage tank.

4. Gravity feed does not establish itself from the main tanks.

5. Cabin air enters the fuel supply manifold and is delivered to all four engines.

6. All four engines eventually flameout.

6.4.1.13.2 Evidence, Data and Rationale: This scenario is corroborated by thecrew’s recorded comments pertaining to fuel flow, the fact that engines were lostsequentially not simultaneously, the incidents believed to be caused by fuel starvationthat were detailed in Section V, the incident history and continuing possibility that maintank pumps can be inadvertently left off, the results of ground testing performed at therequest of the BAR (i.e., Allison’s air injection testing at Little Rock AFB), andthe result of flight testing performed at the request of the BAR (i.e., Air Force testingat Edwards AFB). Although not an exact replication of this scenario was tested at EdwardsAFB, the potential for pressurized cabin air to enter the fuel supply manifold andeventually flameout multiple engines was certainly demonstrated. See Section 7 or thespecific test report document of interest for more details on the results of individualtests.

6.4.1.13.3 This scenario is primarily rebutted by the procedure performed for the usageof fuselage fuel (reference T.O. 1C-130(H)H-1, page 7-7) which incorporates several stepsto ensure fuel is available, pressurized by cabin air and then properly routed to theengines. If this procedure was performed improperly, it is believed that the flightengineer would surely have made the connection with the transition to a new tank and theengine power loss now occurring. With the connection recognized, it is believed the flightengineer would quickly undo that just done and sought another source of fuel.Additionally, we know from the survivor’s testimony that the flight engineer had beenreading a book in the minutes before the power loss occurred. Furthermore, from thecockpit voice recorder, we know that the flight engineer was engaged in a discussion withother crew members at the time when the power loss started. There is absolutely nodiscussion in the cockpit that the crew is about to perform or is performing this atypicalcabin pressurization procedure for moving fuel in a standard environment. Lastly,testimony from flight engineers who knew the mishap flight engineer had never heard himdiscuss a technique where all the main tank pumps would be turned off. All of thesefactors make it unlikely that the use of the cabin pressurization procedure to move fuelresulted in this mishap. Physical evidence needed to further corroborate or rebut thisscenario are:

1. The Wing Section.

2. The Forward Fuel Control Panel and the Auxiliary Fuel Panel.

3. Both Fuselage Tanks.

6.4.1.14 Scenario Number 14. Wrong (i.e., Left Hand) Fuselage Fuel Tank Selected

6.4.1.14.1 Scenario Details. This scenario is a variation of scenario number 13.The fuel burn for this scenario is assumed to be the same as previously described inscenario number 1. The engines fed from their respective main tanks for approximately thefirst 33 minutes of flight. This is followed by 51 minutes of all four engines being fedfuel from the left hand auxiliary tank. At this point, the aircraft is one hour and 24minutes into the flight--the time when the flight engineer needs to use fuel from theright hand fuselage tank and the same time when unexplained torque fluctuations occur. SeeTable 6-4 for fuel-related details about each phase of the sortie for this scenario. Ifthe flight engineer inadvertently turned on the left hand fuselage tank pump(s), insteadof the right hand fuselage tank pump(s), the main tank pumps were turned off and the fuelpanel warning/caution lights not seen or believed, then the engines could have flamed outdue to pressurized cabin air entering the fuel supply manifold via the empty left handfuselage tank. Fuel flow to the engines may have ceased in the following manner:

1. All the main tank pumps are off.

2. Finished crossfeeding from left hand auxiliary tank.

3. Left hand fuselage tank pump selected instead of right hand tank pump.

4. Gravity feed does not establish itself from the main tanks.

5. Cabin air enters the fuel supply manifold and is delivered to all four engines.

6. All four engines eventually flameout.

6.4.1.14.2 Evidence, Data and Rationale: This scenario is corroborated by thecrew’s recorded comments pertaining to fuel flow, the fact that engines were lostsequentially not simultaneously, the incidents believed to be caused by fuel starvationthat were detailed in Section V, the incident history and continuing possibility that maintank pumps can be inadvertently left off, the results of ground testing performed at therequest of the BAR (i.e., Allison’s air injection testing at Little Rock AFB), andthe result of flight testing performed at the request of the BAR (i.e., Air Force testingat Edwards AFB). Although not an exact replication of this scenario was tested at EdwardsAFB, the potential for pressurized cabin air to enter the fuel supply manifold andeventually flameout multiple engines was certainly demonstrated. See Section 7 or thespecific test report document of interest for more details on the results of individualtests.

6.4.1.14.3 This scenario is primarily rebutted by the procedure performed for the usageof fuselage fuel (reference T.O. 1C-130(H)H-1, page 7-7) which incorporates several stepsto ensure fuel is available, pressurized by a pump and then properly routed to theengines. If this procedure was performed improperly, it is believed that the flightengineer would surely have made the connection with the transition to a new tank and theengine power loss now occurring. With the connection recognized, it is believed the flightengineer would quickly undo that just done and sought another source of fuel.Additionally, we know from the survivor’s testimony that the flight engineer had beenreading a book in the minutes before the power loss occurred. Furthermore, from thecockpit voice recorder, we know that the flight engineer was engaged in a discussion withother crew members at the time when the power loss started. Lastly, testimony from flightengineers who knew the mishap flight engineer had never heard him discuss a techniquewhere all the main tank pumps would be turned off. All of these factors make it unlikelythat a closed manual isolation valve resulted in this mishap. Physical evidence needed tofurther corroborate or rebut this scenario are:

1. The Wing Section.

2. The Forward Fuel Control Panel and the Auxiliary Fuel Panel.

3. Both Fuselage Tanks.

6.4.1.15 Scenario Number 15. Synchrophaser Failure

6.4.1.15.1 Scenario Details: For this scenario, the synchrophaser is presumed tohave a single or multiple failure mechanism which results in four-engine flameout. TheC-130 fleet has experienced numerous incidents involving uncommanded engine RPM rollbacksduring flight (reference Section 5). These RPM reductions are characterized as momentaryin nature, usually lasting only several seconds. Typically all four engines are affectedsimultaneously, and the maximum amount of RPM lost in the worst case (i.e., internalsynchrophaser failure) is approximately four percent. There is some ambiguity as to themagnitude of any power loss because the torque gauges in the cockpit are known to provideerroneous readings when their electrical power source is fluctuating. The cockpit torquegauges are powered by the essential AC bus, the same bus powering the synchrophaser.

6.4.1.15.1.1 Fluctuations in essential AC power, essential bus generator failure,faulty/loose synchrophaser-related wiring, exposure of the synchrophaser to moisture,internal synchrophaser failures, or electromagnetic interference can result in four engineRPM rollbacks. All exact failure mechanisms are not known, but the RPM reductions mayoccur if the engines are in the normal or synchrophasing mode and erroneous signals reachthe speed bias motor. The issue addressed by this scenario is whether this situation candeteriorate to engine flameout. The engines may have flamed out in the following manner:

1. An unknown failure mode adversely affects the synchrophaser.

2. Uncontrollable commands are sent from the synchrophaser to all engines.

3. All four engines eventually flameout.

6.4.1.15.2 Evidence, Data and Rationale: This scenario is corroborated by thehistory of synchrophaser induced RPM rollback events in the C-130 fleet. However, sincenone of these events resulted in any engine flameouts, this evidence is not convincing.Additionally, this scenario is corroborated by the aircraft’s history pertaining toelectrical system write-ups. This history includes blown fuses, burned out instrument andpanel light bulbs, several aircraft battery replacements, a broken electrical wire, andthe replacement of two integrated display control units after the morning sortie on 22 Nov96. Analysis of the "failed" integrated display control units indicates failuremay be due to "…temporary periods of poor quality power…"

6.4.1.15.3 This scenario is rebutted by the crew’s recorded comments pertaining tofuel flow, the fact that engines were lost sequentially not simultaneously, and historicalground test results (i.e., Lockheed Report LG88ER0071), and the results of anindependently performed failure modes and effects analysis on the synchrophaser.Additionally, analysis of radar information reveals that there is no traffic close to King56 in the minutes before they experienced problems. This makes it very unlikely thatelectromagnetic interference from another aircraft was the source of their problem. SeeSection 7 or the specific test report document of interest for more details on the resultsof individual tests.

6.4.1.15.4 RPM data obtained from the DFDR reveals steady RPM performance, within thenormal operating range, until torque for each respective engine decays to the point wherethe engine can no longer maintain the propeller at 100% RPM. This is the normal, expectedindication of a properly functioning propeller governing system reacting to a decreasingamount of power available from an engine. Additional evidence rebutting this scenario arethe electrical and mechanical physical limits placed on synchrophaser control authoritywhich precludes engine flameouts from occurring. The FMECA, when complete, will addressmultiple failure modes and effects.

6.4.1.15.5 No additional physical evidence is needed to further corroborate or rebutthis scenario.

6.4.1.16 Scenario Number 16. Temperature Datum System Failure

6.4.1.16.1 Scenario Details: Each temperature datum system has the ability totake and put fuel to its respective engine. Based upon throttle position, specific fuelproperties, and measured engine turbine inlet temperature, each temperature datumamplifier commands its temperature datum valve to take or put fuel to the engine so thattargeted turbine inlet temperature is achieved. It is postulated that fluctuations inessential AC and/or DC power, essential bus generator failure, or electromagneticinterference can result in four engine power losses. All exact failure mechanisms are notknown, but engine power losses may occur if too much fuel is taken from the engines. Inthe extreme case, it is presumed that the temperature datum systems might cause engines toflameout. The engines may have flamed out in the following manner:

1. An unknown failure mode adversely affects all temperature datum systems.

2. The temperature datum systems take too much fuel from the engines.

3. All four engines eventually flameout.

6.4.1.16.2 Evidence, Data and Rationale: This scenario is corroborated by themishap aircraft’s history pertaining to electrical system write-ups. This historyincludes blown fuses, burned out instrument and panel light bulbs, several aircraftbattery replacements, a broken electrical wire, and the replacement of two IDCUs after themorning sortie on 22 Nov 96. Analysis of the "failed" IDCUs indicates failuremay be due to "…temporary periods of poor quality power…"

6.4.1.16.3 This scenario is rebutted by historical ground test results (i.e., LockheedReport LG88ER0071), and the results of ground testing performed at the request of theC-130 BAR (i.e., Allison temperature datum testing at Little Rock AFB). Additionally,analysis of radar information reveals that there is no traffic close to King 56 in theminutes before they experienced problems. This makes it very unlikely that electromagneticinterference from another aircraft was the source of their problem. See Section 7 or thespecific test report document of interest for more details on the results of individualtests.

6.4.1.16.4 This scenario is based on the theory that the engine temperature datumamplifiers may react to a low voltage condition in a manner to reduce fuel flow to theengine, causing the engines to flameout. However, there is no DFDR evidence that theaircraft experienced a low voltage condition. To the contrary, the DFDR indicates noapparent voltage anomaly up to the point when the last engine failed and the lastgenerator dropped off the line.

6.4.1.16.5 Two ground tests have been conducted to see if the temperature datum systemwould react to low voltage and flameout an engine. The first test was conducted byLockheed, Allison, and Hamilton-Standard in 1988. This laboratory instrumented tests of anengine revealed that the temperature datum system was unaffected by low voltageconditions, and the engine did not flameout.

6.4.1.16.6 Recent tests by the Air Force and Allison showed that a specific model oftemperature datum amplifier reduced power output of the engine from 16,000 to 12,000in-lbs of torque when subjected to a low voltage condition. When the fuel control wasmaladjusted to a lean fuel schedule and the temperature datum amplifier was subjected to alow AC voltage condition the torque went from 16,000 to 11,500 in-lbs. 11,500 in-lbs oftorque is still a significant amount of power. A second temperature datum amplifier modeltested did not react to low voltage conditions as in the 1988 test.

6.4.1.16.7 Low voltage conditions did not, in any of the above tests, cause the engineto flameout.

6.4.1.16.8 No additional physical evidence is required.

6.4.1.17 Scenario Number 17. Improper Common Maintenance Action

6.4.1.17.1 Scenario Details: This scenario presumes that there was some common,routine maintenance that was performed on the aircraft that could have adversely affectedall four engines. Examples might include a periodic engine filter replacement wherefilters with too fine a "mesh" were installed. As the aircraft flew, thesefilters became clogged by particulate matter normally allowed to pass through the properfilters, eventually sufficiently restricting fuel flow leading to engine flameouts.Another scenario may include the installation of the proper filters, but their housingswere improperly reassembled leading to nearly simultaneous failure and loss of fuel flow.A final example may include engine oil changes where an inadequate or excessive amount ofoil was added to the engines. The engines may have flamed out in the following manner:

1. A maintenance action, common to all four engines, was performed improperly.

2. All four engines, or the systems providing essential functions/material to sustain combustion, are adversely affected.

3. All four engines eventually flameout.

6.4.1.17.2 Evidence, Data and Rationale: This scenario is corroborated by thecrew’s recorded comments pertaining to fuel flow and the fact that the engines werelost sequentially not simultaneously.

6.4.1.17.3 This scenario is rebutted by the results of the Preflight Inspectionperformed on 21 Nov 96, the results of the Through Flight Inspection performed on 22 Nov96, and the recovery of the number 4 engine. The preflight and thru-flight inspectionsfound nothing significant. Additionally, no common maintenance was performed between themorning and evening sorties on 22 Nov 96. The only significant maintenance performed onthe aircraft on 22 Nov 96 was the replacement of two IDCUs. Lastly, the number 4 engine,which was recovered and analyzed, revealed no evidence of improper maintenance. Physicalevidence needed to further corroborate or rebut this scenario are:

1. Additional Engines.

6.4.1.18 Scenario Number 18. Engine Icing

6.4.1.18.1 Scenario Details: This scenario presumes thaticing conditions were encountered which were either not detected or acted upon. This couldbe due to a failure of either the ice detection or the anti-icing systems. If ice wasallowed to build on the aircraft engines, it is postulated that pieces of ice break free,are ingested by the engines, and flame them out. The engines may have flamed out in thefollowing manner:

1. Icing conditions are encountered.

2. The ice detection system fails or the anti-icing system fails.

3. Unbeknownst to the crew, ice forms on engine inlets.

4. Pieces of ice break free and are ingested by the engines.

5. All four engines eventually flameout.

6.4.1.18.2 Evidence, Data and Rationale: This scenario is corroborated by thefact that the engines were lost sequentially not simultaneously, the forecasted weather,and the fact that there have been other incidents in the C-130 fleet where ice ingestionhas resulted in engine flameouts. During climb, King 56 was told by Seattle Center tolevel off at 15,000 feet for crossing traffic. This level off lasted for approximately twominutes. Light icing was forecast for 10,000 to 17,000 feet on departure. It is unknown ifthe crew activated the engine anti-ice system.

6.4.1.18.3 This scenario is rebutted by the crew’s recorded comments pertaining tofuel flow and the belief that if King 56 experienced icing problems, it would have beenwhile flying at 15,000 feet, not at FL220. FL220 was not forecast for icing. Also, King 56flew approximately 76 minutes after this very brief encounter with forecast icing at15,000 feet. Lastly, the number 4 engine, which was recovered and analyzed, revealed noevidence of internal damage due to ice ingestion. Physical evidence needed to furthercorroborate or rebut this scenario are:

1. Additional Engines.

6.4.1.19 Scenario Number 19. Wrong (i.e., Left Hand) Fuselage Fuel Tank Filled

6.4.1.19.1 Scenario Details: This scenario is a variation of scenario number 14.This scenario assumes that maintenance filled the left hand fuselage tank instead of theright hand one and that all three manual isolation valves are open. The fuel burn for thisscenario is assumed to be the same as previously described in scenario number 1. Theengines fed from their respective main tanks for approximately the first 33 minutes offlight. This is followed by 51 minutes of all four engines being fed fuel from the lefthand auxiliary tank. At this point, the aircraft is one hour and 24 minutes into theflight--the time when the flight engineer needs to use fuel from the right hand fuselagetank and the same time when unexplained torque fluctuations occur. See Table 6-4 forfuel-related details about each phase of the sortie for this scenario. Again, it ispresumed the main tank pumps are off and that there is some crossed wiring such that thefuel in the left hand fuselage tank is indicated on the right hand fuselage tank quantitygauge in the cockpit. When the flight engineer intends to crossfeed "fuel" fromthe right hand fuselage tank in the usual way, pressurized cabin air enters the fuelsupply manifold, via the empty left hand fuselage tank, resulting in four engineflameouts. Fuel flow to the engines may have ceased in the following manner:

1. Fuselage tank quantity indicating wiring is somehow crossed.

2. The left hand fuselage tank is filled instead of the right.

3. All the main tank pumps are off.

4. Finished crossfeeding from left hand auxiliary tank.

5. Right hand fuselage tank pump is turned on.

6. Gravity feed does not establish itself from the main tanks.

7. Cabin air enters the fuel supply manifold and is delivered to all four engines.

8. All four engines eventually flameout.

6.4.1.19.2 Evidence, Data and Rationale: This scenario is corroborated by thecrew’s recorded comments pertaining to fuel flow, the fact that engines were lostsequentially not simultaneously, the incidents believed to be caused by fuel starvationthat were detailed in Section V, the incident history and continuing possibility that maintank pumps can be inadvertently left off, the aircraft’s recent history of fuelgauging problems with some tanks that were being carried as open write-ups, the results ofground testing performed at the request of the BAR (i.e., Allison’s air injectiontesting at Little Rock AFB) and the results of flight testing performed at the request ofthe BAR (i.e., Air Force testing at Edwards AFB). Although not an exact replication ofthis scenario was tested at Edwards AFB, the potential for pressurized cabin air to enterthe fuel supply manifold and eventually flameout multiple engines was certainlydemonstrated. See Section 7 or the specific test report document of interest for moredetails on the results of individual tests.

6.4.1.19.3 This scenario is primarily rebutted by the procedure performed for the usageof fuselage fuel (reference T.O. 1C-130(H)H-1, page 7-7) which incorporates several stepsto ensure fuel is available, pressurized by a pump and then properly routed to theengines. If this procedure was performed improperly, it is believed that the flightengineer would surely have made the connection with the transition to a new tank and theengine power loss now occurring. With the connection recognized, it is believed the flightengineer would quickly undo that just done and sought another source of fuel.Additionally, we know from the survivor’s testimony that the flight engineer had beenreading a book in the minutes before the power loss occurred. Furthermore, from thecockpit voice recorder, we know that the flight engineer was engaged in a discussion withother crew members at the time when the power loss started. Lastly, testimony from flightengineers who knew the mishap flight engineer had never heard him discuss a techniquewhere all the main tank pumps would be turned off. All of these factors make it unlikelythat a closed manual isolation valve resulted in this mishap. Physical evidence needed tofurther corroborate or rebut this scenario are:

1. The Wing Section.

2. The Forward Fuel Control Panel and the Auxiliary Fuel Panel.

3. Both Fuselage Tanks.

6.4.1.20 Scenario Number 20: Synchrophaser and Temperature Datum Failures

6.4.1.20.1 Scenario Details: This scenario is a combination of the synchrophaserfailure and temperature datum system failure scenarios detailed separately as scenariosnumber 15 and number 16. In this case, it is presumed that there is an unknown failuremode such that the combined interaction of both systems reduces both engine RPM and fuelflow in such a manner as to cause all engines to flameout. Another possibility is that thecombined effect would reduce RPM to the point of opening of the acceleration bleed valves,reducing compressor efficiency so that the total effect of all events results in engineflameouts. The engines may have flamed out in the following manner:

1. Unknown failure modes adversely effect the synchrophaser and temperature datum systems.

2. Uncontrollable commands are sent from the synchrophaser to all engines at the same time when the temperature datum systems are taking too much fuel from the engines.

3. As the engine RPM drops to 94%, the acceleration bleed valves open.

4. All four engines eventually flameout.

6.4.1.20.2 Evidence, Data and Rationale: This scenario is corroborated by thehistory of synchrophaser induced RPM rollback events in the C-130 fleet. However, sincenone of these events resulted in any engine flameouts, this evidence is not convincing.Additionally, this scenario is corroborated by the aircraft’s history pertaining toelectrical system write-ups. This history includes blown fuses, burned out instrument andpanel light bulbs, several aircraft battery replacements, a broken electrical wire, andthe replacement of two IDCUs after the morning sortie on 22 Nov 96. Analysis of the"failed" IDCUs indicates failure may be due to "…temporary periods ofpoor quality power…" Lastly, the coupling action of two independent systems(i.e., four temperature datum systems and a synchrophaser) appears even more unlikely eventhough these specific systems receive power from the same AC electrical power source.

6.4.1.20.3 This scenario is rebutted by historical ground test results (i.e., LockheedReport LG88ER0071), and the results of an independently performed failure modes andeffects analysis on the synchrophaser. Additionally, analysis of radar information revealsthat there is no traffic close to King 56 in the minutes before they experienced problems.This makes it very unlikely that electromagnetic interference from another aircraft wasthe source of their problems. See Section 7 or the specific test report document ofinterest for more details on the results of individual tests.

6.4.1.20.4 RPM data obtained from the DFDR reveals steady performance, within normaloperating parameters, until engine torque for each respective engine decays to the pointwhere the propeller governing system can no longer maintain 100% RPM by adjustingpropeller blade angle. This is the normal, expected indication of a properly functioningpropeller governing system reacting to a decreasing amount of power available from anengine. Additional evidence rebutting this scenario are the electrical and mechanicalphysical limits placed on synchrophaser control authority which would preclude engineflameouts from occurring. The FMEA states that no single failure modes exist. The FMECA,when complete, will address multiple failure modes and effects. Physical evidence neededto further corroborate or rebut this scenario are:

1. Additional Engines

Table 6-8. Scenario Summaries Quick Reference

Scenario # and Name

Required Deficiency

History of Required Deficiency

Component or System Test Results

Aircraft Test Results

Additional Wreckage Desired

1

Left Hand Auxiliary Fuel Tank Run Empty 4 MT Pumps off Some GF not a given, Air is Detrimental Torque Flux, No FOs §

2

Right Hand Fuselage Fuel Tank Run Empty 4 MT Pumps off Some GF not a given, Air is Detrimental FOs Occur §

3

Insufficient Fuel Manifold Priming Insufficient Prime Some Air is Detrimental Torque Flux, No FOs §

4

Right Hand Fuselage Fuel Tank Pump(s) Failure 4 MT Pumps off, RH Fus Pump Failure Some, Some GF not a given, Air is Detrimental FOs Occur § and RH Fus Tank Pumps

5

Left Hand Auxiliary Fuel Tank Pump Failure 4 MT Pumps off, LH Aux Pump Failure Some, Some GF not a given, Air is Detrimental No FOs § and LH Auxiliary Tank Pump

6

Undetected Fuel Leak Manifold Leak Some Not Planned Not Planned §

7

Fuel Dump Valve(s) Stuck Open Failed Dump Valve(s) Some Not Planned Not Planned §

8

Refuel/Dump Line Rupture Manifold Rupture None Not Planned Not Planned §

9

Water in Right Hand Fuselage Fuel Tank Water in Fuel Unknown Not Planned Not Planned §

10

Water in Left Hand Auxiliary Fuel Tank Water in Fuel Unknown Not Planned Not Planned §

11

Contaminated Fuel Contam Fuel Some Not Planned Not Planned §

12

Right Hand Fuselage Fuel Tank Manual Isolation Valve Closed 4 MT Pumps off Some GF not a given, Air is Detrimental FOs Occur* §

13

Cabin Pressurization Procedure 4 MT Pumps off Some GF not a given, Air is Detrimental FOs Occur* §

14

Wrong (i.e., LH) Fuselage Fuel Tank Selected 4 MT Pumps off, Wrong Fus Tank Selected Some GF not a given, Air is Detrimental FOs Occur* §

15

Synchrophaser Failure Component Failure Some Torque Flux, No FOs Not Planned None

16

Temperature Datum System Failure Component Failure Some Torque Flux, No FOs Not Planned Additional Engines

17

Improper Common Maintenance Action Repeated Error Unknown Not Planned Not Planned Additional Engines

18

Engine Icing 2 System Failures Some Not Planned Not Planned Additional Engines

19

Wrong (i.e., LH) Fuselage Fuel Tank Filled Wiring Prob, 4 MT Pumps off, Wrong Fus Tank Filled Some, Some, Unknown GF not a given, Air is Detrimental FOs Occur* §

20

Synchrophaser and Temperature Datum System Failure Multiple Component Failures Unknown Torque Flux, No FOs Not Planned Additional Engines

MT = Main Tank; GF = Gravity Feed; FO = Flameout

* Not specifically tested;conclusion based upon similarity to other flight test configurations and results.

§ Wing Section, Forward andAuxiliary Fuel Panels, and Both Fuselage Tanks.

6.5 Summary

6.5.1 Because the foregoing discussion covering all the scenarios is extremely lengthy,a consolidated summary is presented in Table 6-8. In addition, a matrix which succinctlysummarizes which pieces of evidence and data corroborate or rebut each scenario is alsoprovided as Table 6-9. At the bottom of Table 6-9 is a listing of the wreckage componentsneeded to further corroborate or rebut each of the scenarios. Both tables distill atremendous amount of information into very concise, quick references. However, allpertinent information is not included in these tables and they are not intended to besubstitutes for either the detailed scenarios above or the test reports referenced inthem.

6.5.2 For each scenario, the estimated likelihood of occurrence on King 56 iscategorized as either likely or not likely. The determination as to whether a scenario islikely or not likely is based upon physical evidence, ground test results, flight testresults, analysis and professional judgment. Of the 20 scenarios, 16 are categorized asunlikely to have occurred to King 56. These 16 scenarios lack compelling evidence,historical data, or results from testing for their corroboration. In many cases theydepend upon the occurrence of multiple failures and the lack of system redundancy, designsafeguards and crew intervention--these are very unlikely dependencies.

6.5.3 The remaining four scenarios are categorized as likely to have occurred.Scenarios categorized as likely are strongly corroborated by evidence, historical data,and the results of testing. Specifically, logical fuel burn profiles have been developedfor each of these four scenarios suggesting their timing is possible for King 56, and theresults from aircraft testing definitively showed that engine flameouts may be possible ordid occur. In contrast to the 16 unlikely scenarios, there is little rebutting evidence,historical data, or results from testing to suggest that one of these four likelyscenarios did not occur. Specific scenario categorization is as follows:

LIKELY

1. Left Hand Auxiliary Fuel Tank Run Empty

2. Right Hand Fuselage Fuel Tank Run Empty

3. Insufficient Fuel Manifold Priming

4. Right Hand Fuselage Fuel Tank Pump(s) Failure

NOT LIKELY

5. Left Hand Auxiliary Fuel Tank Pump Failure

6. Undetected Fuel Leak

7. Fuel Dump Valve(s) Stuck Open

8. Refuel/Dump Line Rupture

9. Water in Right Hand Fuselage Fuel Tank

10. Water in Left Hand Auxiliary Fuel Tank

11. Contaminated Fuel

12. Right Hand Fuselage Fuel Tank Manual Isolation Valve Closed

13. Cabin Pressurization Procedure

14. Wrong (i.e., LH) Fuselage Fuel Tank Selected

15. Synchrophaser Failure

16. Temperature Datum System Failure

17. Improper Common Maintenance Action

18. Engine Icing

19. Wrong (i.e., LH) Fuselage Fuel Tank Filled

20. Synchrophaser and Temperature Datum System Failure

6.5.4 Of the four scenarios categorized as likely, all are related to fuel managementand the interruption or loss of a constant fuel delivery to the aircraft engines. Ifallowed to persist, it is believed that engine torque fluctuations, followed by flameouts,would occur--the same things that were recorded by King 56’s DFDR. The bestinvestigative efforts to date, utilizing available physical evidence, ground test results,flight test results, analysis and professional judgment, have only narrowed the focus tofour likely scenarios. However, further steps should be taken to further narrow the listof likely scenarios, or add evidence to one of the not likely scenarios to develop onewhich is most probable. To accomplish this in a timely manner and with a high degree ofcertainty, the most prudent step is to obtain more King 56 specific information. That is,recover more wreckage from the ocean floor.

6.5.5 Specific items of recovery should include the wing section, the forward andauxiliary fuel panels, and both fuselage tanks. The wing section should be recovered forits fuel valves and the left hand auxiliary tank fuel pump still contained within it. Theright hand fuselage tank should be recovered for its fuel quantity gauge and two fuelpumps. The forward and auxiliary fuel panels should be recovered for their fuel quantitygauges.

6.5.6 Wing Section. Wreckage recovery video taken by the US Coast Guard clearlyshows the top and bottom of the wing the day that recovery was attempted by the USCGCButtonwood. Based upon this video, we know that specific valves and a fuel pump are stillinside. These valves are DC powered, and upon the loss of the last engine generator, theirpositions are captured. The pump may show signs of failing in flight. An examination ofeach of these components will help narrow the focus to one probable scenario. Lastly, fromthe USCGC Buttonwood’s log and ocean current information, the BAR believes theapproximate wing location is 40.15N and 124.56W. Water depth at this location isapproximately 5,200 feet.

6.5.7 Fuselage Tanks. Wreckage recovery video taken by the US Navy clearly showsboth fuselage tanks. The right hand fuselage tank is of interest because its fuel quantitygauge and fuel pumps will help narrow the focus to one probable scenario. The fuelquantity gauge is AC powered , and upon the loss of the last engine generator, retains itslast reading. From the video, it is difficult to distinguish which of the two tanks is theright hand tank making the recovery of both tanks necessary. From the wreckage video, oneof the fuselage tanks was co-located with the tail section. This location is identified asapproximately 40.08.30N and 125.12.65W in Tab R of the releasable Legal Report. Thelocation of the second tank is less certain, but it is not too far from the first tank.The M/V Laney Chouest recorded views of both tanks on video tape.

6.5.8 Forward and Auxiliary Fuel Panels. The fuel tank quantity gauges fromthese panels are of interest because they will yield clues about which tanks were used assources of fuel sent to the engines during the entire sortie. They will also yield cluesabout how long they were used as fuel sources. In other words, they will validate ordisprove each of the possible fuel burn profiles. Like the fuselage tank fuel quantitygauge, these gauges are AC powered, and upon the loss of the last engine generator, retaintheir last reading. Unfortunately, the location of the forward and auxiliary fuel panelsis not certain. The M/V Laney Chouest recorded exterior views of the cockpit portion ofthe aircraft, but there are not any interior views of the cockpit. Therefore, determiningwhat panels are still present in the cockpit section is currently impossible. As aconsequence, recovery and examination of this cockpit section seems appropriate.Additionally, a comprehensive search of the immediate area surrounding this cockpitsection is necessary as well since the necessary gauges may have broken free. The specificlocation of the cockpit section is not known but the M/V Laney Chouest video taped it on15 Dec 96--the same day it video taped the two fuselage tanks.

6.5.9 It should be noted that although all the fuel quantity gauges work the same way,they are not identical and this will aid in determining a probable scenario even if allthe fuel quantity gauges are not recovered still attached to their respective panels. Forexample, number 1 main tank, number 4 main tank, and both external tank gauges have dialfaces which read from zero to 9000 lbs of fuel. Therefore, recovery of just one of thesegauges, which indicates other than zero lbs, is necessary to get a good idea of the amountof fuel in each of the four main tanks--presuming all four main tanks were used equally.Both external tank gauges should indicate zero lbs since they were not filled with fuelprior to take off. Of course, recovery of additional main tank gauges only provides abetter indication of the fuel load in the main tanks. The number 2 and number 3 main tankgauges have dial faces which read from zero to 8000 lbs; the auxiliary tank gauges havedial faces which read from zero to 6000 lbs; the fuselage tank gauges have dial faceswhich read from zero to 12,000 lbs; and the totalizer (which does not consider fuselagetank fuel in its reading) reads from zero to 68,000 lbs. Given the indications fromseveral gauges, a complete picture of the fuel load on the aircraft can be established.

6.5.10 Once King 56’s wing, fuselage tanks, and fuel quantity gauges arerecovered, the logic tree in Figure 6-10 can be utilized to determine the most probablescenario. The most compelling reason to obtain additional wreckage is the possibility thatevidence which supports an unknown new scenario will be found. Recovery of contradictoryevidence could indicate that a problem exists within the C-130 fleet.

Table 6-9: Scenario Evidence and Data Quick Reference

EVIDENCE & DATA \ SCENARIO #

1

2

3

4

5

6

7

8

9

10

11

12

13

14

15

16

17

18

19

20

EVIDENCE
CVR - General Discussions, Fuel Flow Comments

C

C

CR

C

C

C

C

C

R

R

C

CR

CR

CR

R

~

C

R

CR

~

DFDR - Torque, RPM, Seq v. Sim Flameouts

C

C

C

C

C

C

C

C

C

C

C

C

C

C

R

~

C

C

C

R

Aircraft Records

~

C

~

~

~

~

~

~

~

~

~

~

~

~

C

C

R

~

C

C

Weather

~

~

~

~

~

~

~

~

C

C

~

~

~

~

~

~

~

CR

~

~

Radar Data - Departure & End of Sortie

R

~

~

~

~

~

~

~

~

~

~

~

~

~

R

R

~

~

~

R

Fuel Sample Results

~

~

~

~

~

~

~

~

R

R

R

~

~

~

~

~

~

~

~

~

#4 Engine Teardown Results

~

~

~

~

~

~

~

~

~

~

T

~

~

~

~

T

R

R

~

T

HISTORICAL DATA
Previous Main Tank Pump Switch OFF Events

C

C

~

C

C

~

~

~

~

~

~

C

C

C

~

~

~

~

C

~

Other Previous Related Events

~

~

C

~

~

C

C

~

~

~

C

~

~

~

C

~

~

C

~

C

TEST DATA
1988 LM-Allison-HStd Synchro-TD Test

~

~

~

~

~

~

~

~

~

~

~

~

~

~

R

R

~

~

~

R

1997 Allison TD-Low Voltage Test

~

~

~

~

~

~

~

~

~

~

~

~

~

~

~

R

~

~

~

~

1997 Allison Air Injection Test

C

C

C

C

~

~

~

~

~

~

~

C

C

C

~

~

~

~

C

~

1997 Edwards AFB Ground Test

C

~

~

~

R

~

~

~

~

~

~

~

~

~

~

~

~

~

~

~

1997 Edwards AFB Flight Test

R

C

CR

C

R

~

~

~

~

~

~

C

C

C

~

~

~

~

C

~

1997 Independent Synchrophaser FMEA

~

~

~

~

~

~

~

~

~

~

~

~

~

~

R

~

~

~

~

R

WRECKAGE COMPONENTS \ SCENARIO #

1

2

3

4

5

6

7

8

9

10

11

12

13

14

15

16

17

18

19

20

Floating Wing Section

N

N

N

N

N

N

N

N

N

N

~

N

N

N

~

~

~

~

N

~

Forward & Auxiliary Fuel Panels

N

N

N

N

N

N

N

N

N

N

~

N

N

N

~

~

~

~

N

~

Both Fuselage Tanks

N

N

N

N

N

N

N

N

N

N

~

N

N

N

~

~

~

~

N

~

LH Auxiliary or RH Fuselage Pump(s)

~

~

~

N

N

~

~

~

~

~

~

~

~

~

~

~

~

~

~

~

Additional Engines

~

~

~

~

~

~

~

~

~

~

N

~

~

~

~

~

N

N

~

N

C = Corroborates Scenario
R = Rebuts Scenario
T = To Be Completed
~ = Not Applicable to Scenario
N = Needed to Further Corroborate/Rebut Scenario

Table 6-10: Logic Tree For recovered Wreckage

C-130 Broad Area Review (11)

6.5.11 Related Deficiencies and Concerns

DEFICIENCY: The information in T.O. 1C-130(H)H-1 pertaining to crossfeedoperations is inadequate.

RECOMMENDATION: The BAR supports expanding the wording of the flight manual on page 3-24 to more clearly outline the circumstances when air can be introduced into a fuel supply manifold. The wording should provide guidance recommending discontinuing use of an empty crossfed tank after its tank empty light has flickered for a short period of time. It should provide rationale and guidance stating that there is no compelling reason to ever have three or four main tank fuel pumps turned off during flight. The BAR supports amending all affected C-130 T.O.s accordingly.

DEFICIENCY: In the event of a main tank failure, T.O. 1C-130(H)H-1, page 3-23allows the use of the dump pump from the same main tank to crossfeed to its respectiveengine. This is an adequate procedure but its use should be discontinued before the dumppump inlet is uncovered. According to T.O. 1C-130(H)H-1, page 1-47, this occurs atapproximately 1500 to 2100 lbs, depending upon the specific type of C-130.

RECOMMENDATION: The BAR supports establishment of a fuel quantity limit for use of the above procedure and revising all affected C-130 T.O.s accordingly.

DEFICIENCY: T.O. 1C-130(H)H-1, page 3-24 information pertaining to gravity feedoperations requires improvement.

ON-GOING RESOLUTION: The BAR supports amending the T.O. so as to distinguish between establishing gravity feed and sustaining gravity feed. It should clearly define the recommended operating ceiling for gravity feed operations; T.O. 1C-130(H)H-1 implies 30,000 feet whereas Lockheed Report ER-4120 cites 20,000 feet. It should also address the relative probability and length of time required to establish gravity feed from the main tanks when previous flow was from the auxiliary, external and fuselage tanks. Finally, the T.O. should address the sustainment of gravity feed from the main--that is, what to do and what not to do with the aircraft. The BAR supports amending all affected C-130 T.O.s accordingly.

Section 7.0

Test Summaries Pertinent to the King 56 Mishap

7.1 March 1961 Lockheed Report ER-4120; Full Scale Mockup Test of theC-130B Fuel Feed System

7.1.1 Since the C-130B fuel system incorporated significant design changes from the

C-130A, the new system was evaluated. The results of this evaluation are applicable toKing 56 because the fuel system of an HC-130P, less its fuselage tank refueling system, isvery similar to a C-130B. The C-130B fuel feed system was evaluated on a full-scalelaboratory mockup to determine its capabilities and limitations. The fuel system wastested at simulated altitude, with hot fuel, to determine the capabilities of the entiresystem. A summary of the test results, as quoted from the Lockheed Report, are:

7.1.2 Sea level pressure drop characteristics of the fuel feed system at various flowrates and fuel supply arrangements are presented in this report (meaning the LockheedReport).

7.1.3 Tests indicate that all engine flow is from the auxiliary tank when the crossfeedsystem is fed by all tanks (auxiliary, inboard, and outboard) simultaneously.

7.1.4 Tests during which climb-to-altitude on a hot day was simulated, indicated thatthe engine suction feed system was capable of operating up to altitudes of approximately20,000 feet with the boost pumps shut off.

7.1.5 Scheduled fuel flow with hot JP-4 and aviation gasoline could be maintained atscheduled rate of climb up to 40,000 feet, with tank boost pumps operating; however, arestricted rate of climb schedule must be maintained with aviation gasoline at atemperature of 110° F or higher in order to limit tankinternal pressure.

7.2 April 1988 Lockheed Report LG88ER0071; C-130 Four EnginePower Reduction Test

7.2.1 During the 1980s there were numerous reported power loss events (reference Figure5-3) which ultimately culminated in power loss testing performed in early 1988. Theresults of this test effort are documented in Lockheed Report LG88ER0071. Selectedparagraphs from the report’s Foreword and Summary are presented below.

7.2.2 Four engine power reductions have been experienced on the C-130 with increasedregularity. Investigations of aircraft incidents indicated that approximately 75% of thepower reductions were caused by failures in the AC power system essential bus, with theremainder caused by failure in the synchrophasing system.

7.2.3 This test was devised to demonstrate the power reduction, to determine whatportion of the engine and propeller controls was being affected when the essential buswould experience failures that caused the voltage to vary over a wide range, and toexamine possible solutions that would prevent the power reduction.

7.2.4 The test set up included a single T56-A-15 engine on a test stand which wasmodified to record all vital data pertaining to the engine and propeller while the engineran as a master or a slave engine. Tests were run using a tube-type synchrophaser and asolid-state synchrophaser with the capability to vary the voltage individually orsimultaneously to the synchrophaser, temperature datum amplifier, and the engineinstruments.

7.2.5 The test was conducted on February 24-25 and March 10, 1988, by running a totalof 144 separate test cases. The following observations were drawn from the data takenduring the tests:

7.2.5.1 The maximum torque loss was 32.4% (5,200 inch-pounds) when the tube-typesynchrophaser was used and the voltage was varied from 115 to 50 and back to 115 VAC usinga 5-second ramp time.

7.2.5.2 The torque loss using the solid-state synchrophaser was of lesser magnitude.The worst case for the solid-state showed a torque loss of 11.8% (1,600 inch-pounds) whenthe voltage was varied from 115 to 0 to 115 using a 3-second ramp time. The torque losswas even less, 2.9%, when the voltage was varied the same as the tube-type.

7.2.5.3 The temperature datum system does not contribute to the torque loss when thevoltage is varied as low as 30 VAC. However, when the voltage went all the way to 0 VAC,the TD system reverts to the hydromechanical fuel schedule (null) and a torque variationof 9.3% (1,260 inch-pounds) was detected.

7.2.5.4 No torque loss could be induced when the propeller governor control was placedin mechanical governing except as described in No. 3 above.

7.2.5.5 At flight idle power setting (approximately 4,000 inch-pounds), no torque losswas noted with either the tube-type or solid-state synchrophaser.

7.2.5.6 Using the Hamilton Standard time delay relay, the magnitude of the torquereduction was limited to 4.4% with the tube-type synchrophaser.

7.2.5.7 With the constant voltage transformer installed, no torque loss occurred.

7.3 Engine Testing - Conducted at Allison in 1988

7.3.1 The "Rollback" tests conducted in 1988 at Allison, on a highlyinstrumented test stand, concluded:

1. The synchrophaser is sensitive to voltage variations and can command propeller pitch changes during an electrical power system malfunction.

2. The torquemeters are also sensitive to voltage variations, and are therefore unreliable indicators of actual torque during an electrical power interruption.

3. The TD amplifier system does not contribute to four-engine power fluctuations.

7.4 October 1997 Allison Temperature Datum Low Voltage

7.4.1 Tests were conducted at Little Rock AFB, AR, to see how an engine would react toreductions in voltage to the Temperature Datum Control system (TD Amplifier).

7.4.2 The TD system operates on both Alternating Current (AC 115 Volts, 400HZ), andDirect Current 24-28 Volts DC. Two amplifiers were tested, one manufactured by Raven, oneby Allied Signal Bendix. Both were solid state amplifiers, which are the current modelsused in the fleet.

7.4.3 Summary of Test Results:

1. The Raven TD Amplifier was affected by AC voltage drops. The Power level in the engine went from 16,000 in-lbs, to 12,000 in-lbs, temperature went from 963° C to about 855° C, due to the AC voltage dropping slowly from 115 volts AC, down to approximately 48 volts AC.

2. When the Fuel Control was leaned out 20° C, (a worst case condition), the power level dropped from 16,000 in-lbs to 11,500 in-lbs, and the temperature went from 967° C to 843° C and the RPM dropped from 99.4% to 94.5% RPM when the AC voltage was dropped quickly from 115 volts AC to 48 volts AC. Also, with the engine in this low power condition the DC voltage was dropped, and the Raven TD Amplifier "LOCKED" the system at a low power setting. This is the "worst case condition".

3. The Raven model TD Amp did not react to a DC only voltage drop, engine power, RPM, were unaffected.

4. The Bendix TD Amplifier was not affected by AC or DC voltage drops, and therefore did not affect engine power or RPM.

5. A DC voltage drop eventually puts the TD valve in the "LOCKED" condition, and the engine anti-ice system activates.

7.4.4 Conclusion: Voltage drops in both AC and DC affect the TD system depending onwhich model TD amplifier is used, and can cause power reduction in the engine. But, thisdoes NOT cause the engine to "Flameout" or quit.

7.5 November 1997 Allison Air Injection Test (Test Stand)

7.5.1 Air was introduced into the fuel supply line for an engine on a test stand atLittle Rock AFB. Fuel delivery pressure to the engine was regulated to 25 psi for alltests. The fuel drain was removed from the fuel heater/strainer and replaced withappropriate number 4, or 1/4 inch, fittings so that air could be introduced into thesystem.

7.5.2 As the supply air pressure was increased from zero to 20 psi, no change in engineoperation was noted. As the supply air pressure was increased to 30 psi, fuel flow, TIT,torque and then RPM started to decay. Air pressure was reduced to zero psi and the enginerecovered.

7.5.3 Supply air pressure was increased from zero to 26 psi and then held at 26 psi,until the engine stabilized. Fuel flow, TIT, and torque decayed significantly, but notRPM. Air pressure was reduced to zero psi and the engine recovered again.

7.5.4 The fuel flow, torque, and TIT are affected by air induced into the fuel systemat pressures as low as one psi above fuel boost pressure.

7.6 November 1997 Edwards AFB Testing

7.6.1 HC-130 fuel starvation testing was accomplished to determine the AllisonT56-A-15 turboprop engine response, as installed, to various conditions which could causeinadequate fuel flow and/or introduce air into its fuel supply lines. The specific testobjectives were:

7.6.2 Auxiliary Tank Pump Failure -- With the engines under test crossfeedingfrom the left hand auxiliary tank, and their respective main tank pumps turned OFF, a lefthand auxiliary tank pump failure was simulated by turning the left hand auxiliary tankpump OFF.

7.6.3 Auxiliary Tank Empty -- With the engines under test crossfeeding from theleft hand auxiliary tank, and their respective main tank pumps turned OFF, the left handauxiliary tank was run empty.

7.6.4 Fuselage Tank Pump Failure -- With the engines under test crossfeedingfrom the right hand fuselage tank, and their respective main tank pumps turned OFF, aright hand fuselage tank pump failure was simulated by turning the right hand fuselagetank pump OFF. This test objective was accomplished with the manual isolation valve on theempty left hand fuselage tank both opened and closed.

7.6.5 Fuselage Tank Empty -- With the engines under test crossfeeding from theright hand fuselage tank, and their respective main tank pumps turned OFF, the right handfuselage tank was run empty.

7.6.6 External Tank Pump Failure -- With the engines under test crossfeedingfrom an external tank, and their respective main tank pumps turned OFF, an external tankfuel pump failure was simulated by turning the external tank pump OFF.

7.6.7 External Tank Empty -- With the engines under test crossfeeding from anexternal tank, and their respective main tank pumps turned OFF, the external tank was runempty.

7.6.8 Improper Priming -- With the engines under test crossfeeding from the lefthand auxiliary tank, and their respective main tank pumps turned OFF, a transition tocrossfeeding from the right hand fuselage tank was performed. The fuel supply manifoldswere not primed before transitioning to the right hand fuselage tank.

7.6.9 Improper Priming (Boost pumps ON) -- With the engines under testcrossfeeding from the left hand auxiliary tank, and their respective main tank pumpsturned ON, a transition to crossfeeding from the right hand fuselage tank was performed.The fuel supply manifolds were not primed before transitioning to the right hand fuselagetank.

7.6.10 Gravity Feed Evaluation -- With the engines under test crossfeeding fromthe left hand auxiliary tank, and their respective main tank pumps turned OFF, the maintank crossfeed valves for the engines under test are switched from open to closed.

7.6.11 Left Bank Evaluation -- With the three engines under test torquefluctuating below 4000 in-lbs simultaneously, a standard rate left hand turn wasinitiated. Torque fluctuations were established by a right hand fuselage tank pump failure(with the manual shutoff valve for the left hand fuselage tank open) and a right handfuselage tank being run empty.

7.6.12 Recovery Procedure Evaluation -- Determine a torque fluctuating andflamed out (caused by fuel starvation) HC-130 engine response to the following recoveryprocedure: (1) Main tank boost pumps ON, (2) Main tank crossfeed valves closed.

7.6.13 The aircraft DFDR was utilized to capture data during all of the testobjectives. To augment this data collection, equipment was added to the test aircraft torecord turbine inlet temperatures, fuel flows, and fuel pressures for each of the engines.As a backup, the engine instrument panel was also video taped during ground and flighttesting.

7.6.14 Ground and flight testing was accomplished at Edwards AFB between 5 and 24 Nov97 on HC-130(H)N, 90-2103. Testing followed a build up approach. Ground testing wasconducted first beginning with single engine scenarios and progressing through four enginescenarios. Single engine tests were conducted on engine number 1, two-engine tests onnumber 1 and number 2, and three-engine tests on number 1, number 2, and number 4. A"flameout" was defined as RPM rolling below 94%. In general, the single engineand 2-engine scenarios were allowed to proceed to flameout if they began fluctuating. The3-engine cases were generally allowed to continue until torque had dropped to 4000 in-lbson all three engines prior to recovery.

7.6.15 Ground tests were conducted on a level engine-run pad at Edwards AFB(approximately 2300 ft pressure altitude) with engines set at 970°C turbine inlet temperature (970 TIT) which produced approximately 15,000 in-lbs of torqueand 1,900 pounds per hour (pph) of fuel flow per engine. Flight tests were conducted at22,000 ft pressure altitude (PA) with engines set at 970 TIT during straight and levelunaccelerated flight except for the left bank evaluation. A left bank evaluation wasconducted at 22,000 ft PA with engines set at 910 TIT. All flight tests were conductedwith an aircraft differential pressure of approximately 7.4 psi.

7.6.16 No fuel starvation scenarios produced engine flameout on the ground. However,the fuselage and external tank pump failure and tank running empty scenarios causedsignificant engine power loss and flameouts in flight. The results of both ground andflight test results are summarized in Table 7-1. Following the table are more detailedexplanations of the results for each test objective.

Table 7-1: HC-130H(N) S/N 90-2103

HC-130 Fuel Starvation Test Results SUMMARY

GROUND TEST

FLIGHT TEST

Total Number of Engines Tested Total Number of Engines Tested

Objective

1

2

3

4

1

2

3

1. Auxiliary Tank Pump Failure

NR

NR

NR

NR

NR

NR

NR

2. Auxiliary Tank Empty

fx

fx

fx

FX

NR

NR

NR

3. Fuselage Tank Pump Failure

NR

NR

NR

NR

FO

FO

FO

4. Fuselage Tank Empty

NR

NR

NR

NR

FO

FO

FO

5. External Tank Pump Failure

~

~

~

~

FO

FO

FO

6. External Tank Empty

~

~

~

~

FX

FO

FO

7. Improper Priming

NR

NR

FX

FX

fx

FX

NR

8. Improper Priming (BP’s ON)

~

~

~

~

~

~

NR

9. Gravity Feed Evaluation

~

~

~

~

NR

~

~

10. Left Bank Evaluation

Not Conclusive

11. Recovery Procedure Evaluation

All Recovered

NR - No significant engine response fx - Torque fluctuations of 1000-4000in-lbs

FX - Torque fluctuations greater than 4000 in-lbs FO - Engine(s) flamedout

~ - Not tested

7.6.17 Auxiliary Tank Pump Failure -- The auxiliary tank pump failure scenariodid not cause abnormal engine response. The fluctuations which did occur during testing ofthis objective were small enough to be within normal engine and gauge fluctuation,especially during operation on the ground. The fact that no abnormal engine responseoccurred indicates that the engines smoothly transitioned from crossfeeding from the leftauxiliary tank to gravity feeding from their main tank.

7.6.18 Auxiliary Tank Empty -- The auxiliary tank running empty scenario causedsignificant torque, TIT, and fuel flow fluctuations on the ground but not in flight.Engine fluctuations from this scenario were greatest on the number 2 engine and grew inmagnitude as testing progressed in number of engines during ground and flight test. Theduration of engine number 2’s fluctuations lasted approximately 50 seconds duringground test and 30 seconds or less in flight. Engine response was significantly larger onthe four engine ground test than the three engine ground test. Since four engine flighttests were not conducted, it is unknown whether this scenario would have producedsignificant engine power loss or flameout. However, significant engine power loss orflameout from the auxiliary tank running empty may be possible.

7.6.19 Fuselage Tank Pump Failure -- The fuselage tank pump failure scenariocaused engine flameouts in flight. During three engine flight tests of the fuselage tankpump failure scenario, engine number 4 experienced torque, TIT, and fuel flow fluctuationsfirst followed by engines number 1 and number 2 within 30 seconds.

7.6.19.1 The fuselage tank pump failure flameouts were unexpected test results, becausecabin pressurization was expected to feed the engines with the fuselage tank pump failed.The cause of the flameouts were thought to be pressurized cabin air which entered throughthe empty left fuselage tank and reached the engines. In order to test this theory, a twoengine fuselage tank pump failure scenario was flight tested with the left fuselage tankmanual isolation valve CLOSED. Closing the manual isolation valve was expected to isolatethe empty tank and block the source of air causing flameout. When the manual isolationvalve to the left fuselage tank was CLOSED, the fuselage tank pump failure scenario causedsignificant engine torque, TIT, and fuel flow fluctuations. The maximum fluctuationsobserved were approximately 12,000 in-lb of torque, 400 degrees of TIT, and 1,100 pph offuel flow. These fluctuations lasted for approximately 2 minutes and 30 seconds before theengines recovered to full power crossfeeding from the right fuselage tank with the rightfuselage tank boost pump OFF (cabin pressurization acting as a fuel pump). The cause ofthe significant engine response when the empty fuselage tank’s manual isolation valveis CLOSED is thought to be pressurized cabin air entering through the right fuselagetank’s vent valve.

7.6.19.2 The engines reverted to gravity feed during ground test of this scenario whenthe aircraft was not pressurized, but the engines did not establish gravity feed inflight. This also supports the theory that pressurized cabin air caused engine flameoutsby blocking main tank fuel from gravity feeding.

7.6.20 Fuselage Tank Empty -- The fuselage tank running empty scenario causedengine flameouts in flight. During three engine flight tests of the fuselage tank runningempty scenario, engines number 1, number 2, and number 4 experienced near simultaneoustorque, TIT, and fuel flow fluctuations. Engine power loss and flameout occurred morequickly than in the fuselage tank pump failure scenarios, since a larger volume of aircould enter fuel lines through an empty tank than through the tank vent valve. Again,flameouts may be caused by pressurized cabin air entering the crossfeed manifold from theempty fuselage tank and blocking main tank fuel from gravity feeding.

7.6.21 External Tank Pump Failure -- The external tank pump failure scenariocaused engine flameouts in flight. The external tank scenarios were not ground testedbecause they were added to the test plan after ground test was complete. During threeengine flight tests of the external tank pump failure scenario, engine number 2experienced significant torque, TIT, and fuel flow fluctuations approximately two minutesand 30 seconds before engines number 4 and number 1 followed.

7.6.21.1 The external tank pump failure flameouts were unexpected test results, becausethe external tanks were expected to act like the auxiliary tanks since both tanks arevented to atmospheric pressure. However, when feeding from the external tanks thecrossfeed manifold is connected to the dump manifold through a one way check valve. Thedump manifold is connected to the fuselage tanks which are at cabin pressure. Therefore,when feeding from external tanks, pressurized cabin air can force its way to the enginesthrough a one way check valve if the pressure in the crossfeed manifold drops low enough.This theory is supported by the fact that engine inlet pressure slowly rose aboveatmospheric pressure as the engines experienced power loss leading to flameouts. Thisindicates that pressurized cabin air was slowly forcing its way into the crossfeedmanifold, blocking main tank fuel from gravity feeding and flaming out engines.

7.6.22 External Tank Empty -- The external tank running empty scenario causedengine flameouts in flight. During three engine flight tests of the external tank runningempty scenario, engine number 2 experienced significant torque, TIT, and fuel flowfluctuations approximately one minute and 30 seconds before engines number 4 and number 1followed.

7.6.22.1 The external tank running empty flameouts were unexpected test results,because the external tanks were expected to act like the auxiliary tanks since both tanksare vented to atmospheric pressure. However, when feeding from the external tanks thecrossfeed manifold is connected to the dump manifold through a one way check valve. Thedump manifold is connected to the fuselage tanks which are at cabin pressure. Therefore,when feeding from external tanks pressurized cabin air can force its way to the enginesthrough a one way check valve if the pressure in the crossfeed manifold drops low enough.This theory is supported by the fact that engine inlet pressure slowly rose aboveatmospheric pressure as the engines experienced power loss leading to flameout. Thisindicates that pressurized cabin air was slowly forcing its way into the crossfeedmanifold, blocking main tank fuel from gravity feeding and flaming out engines.

7.6.23 Improper Priming -- Improper priming scenarios caused severe but briefengine responses during ground and flight testing. Air introduced into the crossfeedmanifold caused large rapid engine power loss (to zero torque) during one and two engineflight tests. Engine fluctuations lasted less than 10 seconds as the engines recoveredwhen pressurized fuel from the fuselage tank boost pump reached the engines. A moreoperationally representative procedure which introduced air between the fuselage tanks andthe right external crossfeed valve did not produce any engine response during three engineflight test.

7.6.24 Improper Priming (Main Tank Boost Pumps On) -- The improper priming withmain tank boost pumps ON scenario did not cause any abnormal engine response.

7.6.25 Gravity Feed Evaluation -- The gravity feed evaluation procedure causedno abnormal engine response. The fact that no abnormal engine response occurred indicatesthat the engines smoothly transitioned from crossfeeding from the auxiliary tank togravity feeding from their main tank.

7.6.26 Left Bank Evaluation -- Response of torque fluctuating engines to pitch,bank, and normal acceleration was inconclusive.

7.6.27 Recovery Procedure Evaluation -- This objective was tested concurrentlywith all other test objectives. The recovery procedure was used for every torquefluctuating engine recovery, after the test point success criteria were complete, as partof the test point clean up procedure. In the case of the torque fluctuating engines (4000in-lbs or less total torque), the main tank boost pumps for the engines under test wereturned on and then the engine crossfeed valves were closed. In the case of the inboardflamed out engine (windmilling at 45% RPM), the boost pump was turned on then thecrossfeed valve closed. The outboard flamed out engine was recovered as RPM was droppingbelow 94%. No attempt was made to recover multiple engines with a single main tank boostpump; however, based on the rapid rate that the crossfeed manifold was pressurized and therapid recovery of the engines, it should be possible to do so.

7.6.27.1 Engines recovered from torque fluctuating conditions within five seconds aftertheir main tank boost pumps were turned ON. When the procedure was conducted on flamed outengines, engine start occurred more slowly. The longest time noted was 20 seconds betweenmain tank boost pump ON and engine number 2 start when engine number 2 was flamed out androtating at approximately 45% RPM.

7.6.27.2 Overall, 22 successful recoveries of torque fluctuating engines wereaccomplished in 22 attempts and seven successful recoveries of flamed out engines wereaccomplished in seven attempts. In every case, the engines recovered to normal operationfrom any of the fuel starvation scenarios tested when they received positive fuel pressurefrom their respective main tank boost pump.

7.6.28 From this data, the C-130 BAR concludes that several scenarios involving thefailure of pumps in selected crossfed tanks with the main tank pumps OFF, or runningselected tanks empty with the main tank pumps OFF have the potential to flameout fourengines in flight.

7.7 December 1997 Allison Air Injection Test (Aircraft)

7.7.1 Air was introduced into the fuel supply line for an engine on an aircraft atLittle Rock AFB. Fuel was gravity fed from the tank to the engine. The fuel drain wasremoved from the fuel heater/strainer and replaced with appropriate number 4, or 1/4 inch,fittings so that air could be introduced into the system.

7.7.2 The air pressure was increased from 0 to 3 psi, resulting in engine powerfluctuations and minimum performance parameters of 94.5% engine RPM, 700° C TIT, 6,500 in/lbs torque, and 1050 lbs per hour fuel flow.

7.7.3 The fuel flow, torque, TIT, and RPM are affected by air introduced into the fuelsystem at pressures as low as three psi when fuel is being gravity fed.

7.8 Innovative Technologies Report, Synchrophaser FMEA

7.8.1 A Task Order was awarded to Innovative Technologies Corporation (ITC) under theWR-ALC/LB EASES contract to conduct a rigorous study of critical C-130 aircraft systems todetermine what possible failures could lead to the loss of all four engines. Specifically,four systems (Electrical System, Fuel System, Engine Control System, and the PropellerControl System) are to be investigated with emphasis of the HC-130P configuration. Inaddition, a separate Failure Modes, Effects and Criticality Analysis (FMECA) is to beperformed on the Synchrophaser subsystem of the Propeller Control System.

7.8.2 The study will be performed in two phases with a possible third phase. Phase 1 isthe effort to scope the overall analysis, conduct a preliminary (quick look) analysis andsubmit a preliminary FMEA of the synchrophaser subsystem and prepare a detailed projectplan. Phase 2 is the performance of a detailed analysis of the four systems to includeFailure Modes, Effects and Criticality Analysis, Fault Tree Analysis, Sneak CircuitAnalysis, Electromagnetic Interference (EMI) Analysis and simulation to the degreeappropriate to determine failures that could cause loss of all four engines. At the end ofPhase 2, a detailed report will be prepared and submitted on the systems analyzed toinclude recommendations on any further analysis required to identify the four enginefailure cause(s). A final FMECA on the Synchrophaser sub-system is also to be conductedand submitted. Phase 3 would be comprised of any additional analysis and any other testsrequired or identified in Phase 2, and the submittal of a final report.

7.8.3 The analysis effort was assigned to members of the ITC team capitalizing upon theexpertise of each team member. ITC is the prime contractor performing as the team lead andintegrator of the effort. Also, ITC performed the analysis of the Fuel and ElectricalSystems. Mercer Engineering Research Center (MERC), a subcontractor to ITC, performed theanalysis of the engine control and propeller control systems. During Phase 1, ScienceApplications International Corporation (SAIC), another ITC sub-contractor, identifiedspecial analysis tools to be employed during the Phase 2 analysis.

7.8.4 No single failure was identified in any of the four systems investigated in thispreliminary analysis that could result in loss of all four engines other than running outof fuel. This is due to a conscious design approach for all systems to provide redundancyas required to prevent a catastrophic event associated with a single failure. Furtheranalysis will most likely have to consider off-nominal performance and/or multiplefailures to find event(s) that could lead to four-engine failure.

Section 8.0

C-130 Fleet Safety Corrective Actions

The team identified a number of corrective actions it felt were necessary to implementits review and appropriate for improving C-130 flight safety overall. Our list of actionsfollows:

8.1 Complete

8.1.1 Publish a Flight Crew Information File entry (known as the "FCIF," asource of immediate information essential for crews to review before their next flightfrom home station), reminding flight crews to adhere to flight manual procedures forleaving main tank boost pumps on in flight. The BAR took this action to correctmisconceptions the BAR had encountered that there were advantages to operating with themoff--there were none under normal operating conditions.

8.1.2 Publish a message telling everyone within the Air Force to report all enginepower-loss incidents as at least a "HAP" (High Accident Potential). The BARimplemented this procedure to enable proper tracking of engine power-loss incidents forfurther analysis as required.

8.1.3 Publish a message reiterating instructions on handling procedures for digitalflight data recorders. This was done to improve the chances for gathering critical flightdata for analysis in future incidents.

8.1.4 Establish and publish a toll-free number available to all personnel for reportingC-130 engine power-loss incidents. The BAR put this number into operation to gather asmuch information as possible about prior incidents, and to identify possible scenarios forevaluation.

8.1.5 Publish by message the accepted definitions of the terms "power-loss,""rollback," and "flameout." The BAR did this to establish clarity indiscussions involving the BAR.

8.1.6 Conduct tests on the Allison temperature datum system on the engines powering theC-130 fleet.

8.1.7 Publish a critical action "bold face" procedure to address the serioussituation of multiple engine power loss/RPM rollback emergencies. Results of BAR initiatedflight tests have confirmed the inability of the engines to sustain combustion due to fuelstarvation in certain situations. These situations can result in power losses that can berecovered by turning on the main fuel tank boost pumps and closing the crossfeed valves.If no corrective action is taken all four engines may flameout.

8.1.8 C-130 ground and flight tests. These tests reviewed a number of scenarios andattempted to duplicate conditions which might have led to the fuel starvation of theengines on King 56. They looked at variations in operating procedures (i.e., changes inswitch positions, boost pump operation, pressurization, etc.), which could havecontributed to the incident.

8.1.9 Allison T56A-7B engine aeration tests (with engine on a test stand) at LittleRock Air Force Base. The ground tests showed that fuel flow, torque, TIT, and RPM areaffected by air induced into the fuel system at pressures as low as one psi above fuelboost pressure.

8.1.10 Allison T56A-7B engine aeration tests (with aircraft-mounted engine) at LittleRock Air Force Base. The ground tests showed that fuel flow, torque, TIT, and RPM areaffected by air introduced into the fuel system at pressures as low as three psi when fuelis being gravity fed.

8.2 In Progress

8.2.1 The C-130 "Tiger Team." This broad, multi-disciplinary study by the AMCstaff and representatives from the Air National Guard, Air Force Reserve Command, and theseveral operating commands, is looking at several aspects of the Air Force’s C-130fleet, training program, personnel, and their future. The Tiger Team’s reportprovides recommendations for the future of the C-130 fleet, how it should be based andemployed, and what should change in C-130 training and operations in the future.

8.2.2 The C-130 BAR. Completion of this report concludes the work of the review.

8.2.3 C-130 Failure Modes, Effects and Criticality Analysis. The FMECA results areexpected in November of 1998 and should provide a detailed summary of those conditionsunder which the C-130’s engines are most likely to fail, and how those failures mightoccur and be prevented.

8.2.4 Warner-Robins continue to evaluate Forms 1067 suggesting constant ignitionsource, reverse current relay rewire, and synchrophaser wiring harness replacement.

8.2.5 Warner-Robins continue to purchase and install a form, fit, and functionreplacement digital flight data recorder (DFDR) for the C-130, and to appropriately"hot-wire" it from the battery, including a "g-cutoff" switch todiscontinue operation after impact.

8.3 Future

8.3.1 The Air Force should review and update the existing lead command operatinginstruction to:

1. Fully reflect command changes which have occurred since the airlift C-130 fleet transferred from Air Combat Command to Air Mobility Command.

2. More fully define the lead command’s leadership role, its responsibilities (particularly with respect to configuration control), its authority to enforce configuration control, and the accountability of other commands to the lead’s direction.

3. Empower the lead command and properly resource the lead and other user/ supporting commands to enable them to:

a. Update, consolidate and standardize aircraft technical manuals and operating guidance to assure crews have standardized procedures and correct performance data.

4. Do the same for maintenance manuals to assure maintainers have the up-to-date information they need to properly maintain the aircraft.

5. Set and enforce a limit on the number of change pages introduced into a manual before it must be replaced, and require emergency procedure changes to be distributed in printed replacement pages within not more than 72 hours of release to eliminate write-in changes.

8.3.2 The Air Force should consider the Federal Aviation Administration and NationalTransportation Safety Board guidelines and directives when arriving at a standardized setof digital flight data recorder flight parameters which the Air Force should thenincorporate into its existing and planned future weapon systems. This would insure thatessential flight data is captured for evaluation in future incidents and accidents. Therecommendations made previously by the Defense Science Board should be included as a partof this effort.

8.3.3 Until they are replaced with newer, more capable models of the C-130, the AirForce should reevaluate and more closely monitor the EC-130 Commando Solo II mission. Theaircraft have the highest empty gross weights of the fleet, owing to the broadcastequipment they carry. The relative age of the aircraft, the requirement to refuel to nearemergency gross weight limits for deployments on operational missions, and the highpotential for Radio Frequency Interference (RFI) induced electrical problems, combine toidentify this mission as one associated with marked higher risk than others. One specificsafety concern the team has is that, despite limited two-engine-out sustained flightcapability, crewmembers routinely carry no parachutes. The BAR believes this should alsobe addressed. The 193rd Special Operations Wing at Harrisburg, Pennsylvania isslated to receive new aircraft eventually, subject to defense budget constraints. Untilthen, the unit must continue to very carefully balance the performance limitations of itsaircraft with the demands of its operating environment. The BAR recommends continued closeoversight by the unit’s higher headquarters to make certain this balance is safelymaintained.

8.3.4 The Air Force should bring the initial mission qualification and continuationtraining for crew members under the review and standardization of C-130 formal training toassure that all crew members are taught and evaluated on the same procedures. This willaid in eliminating misconceptions about deviations from these procedures, and will help inquickly spreading the word when a procedure needs to be changed as a result of newinformation.

8.3.5 The Air Force should bring all initial qualification training students to C-130formal training for the same reason--standardization of C-130 operating proceduresworldwide.

8.3.6 The Air Force should continue the installation of a second pump in the fuselagetanks for these and other similarly equipped aircraft.

8.3.7 The Air Force should establish standardized requirements for aircrews to reviewditching and bailout on the first leg of each over-water mission, given the importance ofcrews maintaining a reasonable level of familiarity with these procedures. As part of thiseffort, the Air Force should:

1. Conduct an analysis of world-wide ditching events. That data should be used to update and standardize all flight manuals with an accurate discussion of ditching survivability and techniques.

2. Review the information concerning bailout in the flight manuals for consistency between models of the same aircraft, and revalidate the accuracy of the information provided to the crew.

8.3.8 Once officially released by the Air Force, incorporate lessons learned from theKing 56 incident into a training video presentation for reference and review by all C-130crew members. An updated training video would incorporate the lessons learned from King56, the Colombian ditching, the gunship mishap in Africa, and other ditching events, andwould be a valuable training tool.

8.3.9 The Air Force should develop and implement a modification to replace thesynchrophaser interface wiring bundles on all C-130 aircraft.

8.3.10 The C-130 SPO should develop and implement a modification which would providepower to the DFDR and CVR during battery only operation.

8.3.11 The BAR recommends renewed Air Force-wide emphasis on fuel sampling as part ofstandard maintenance operations to help identify the presence of water or othercontaminants in the fuel.

8.3.12 The BAR recommends renewed Air Force-wide emphasis on proper testing of theC-130 synchrophaser following reported malfunctions, and on more thorough training formaintenance personnel in the performance of those tests.

8.3.13 The BAR recommends that the Air Force establish a fuel quantity limit for usingthe dump pump as an alternate source of fuel pressure in the event the main boost pumpfails and revision of all affected C-130 T.O.s accordingly.

8.3.14 The BAR supports expanding the wording of the flight manual on page 3-24 to moreclearly outline the circumstances when air can be introduced into a fuel supply manifold.The wording should provide guidance recommending discontinuing use of an empty crossfedtank after its tank empty light has flickered for a short period of time. It shouldprovide rationale and guidance stating that there is no compelling reason to ever havethree or four main tank fuel pumps turned off during flight. The BAR supports amending allaffected C-130 T.O.s accordingly.

Section 9.0

Appendices

9.1 Glossary of Terms

Term Definition

ABCCC Airborne Command, Control, and Communications. A flying battlefield operations command post.
ACC Air Combat Command. The major Air Force Command over attack, fighter, rescue, and bomber aircraft.
Aerial Delivery Combat Air Delivery wherein cargo or paratroopers are dropped by parachute.
AFI Air Force Instruction
AFSOC Air Force Special Operations Command. The major Air Force Command over special air operations (PSYOP, gunship, ABCCC, etc.)
Airland Combat Air Delivery wherein cargo or personnel are delivered by landing at an airfield.
AMC Air Mobility Command. The major Air Force Command over cargo aircraft and airlift operations.
ANG Air National Guard
APU An AC Auxiliary Power Unit
Assault Landing A short field landing (3,500’ or shorter runway).
AWADS Adverse Weather Aerial Delivery System
Bailout Emergency egress of an aircraft via parachute.
BAR Broad Area Review
Bold Face Critical items in the flight manual that must be performed immediately to avoid aggravating an emergency and causing injury or damage. These items must be committed to memory.
CAD Combat Air Delivery - delivering cargo by landing at an airfield (airland) or dropping it by parachute (aerial delivery).
Class A Mishap An accident which results in loss of life, permanent disability, loss of aircraft, or over $1M in damage.
Commando Solo II The name for the Psychological Warfare (PSYOP) Mission flown by one ANG unit.
CRM Cockpit Resource Management. Training which focuses on enhancing crew synergy and coordinating mission accomplishment and handling unusual or emergency situations.
Crossfeed To transfer fuel from any tank by routing the fuel through the crossfeed manifold.
CVR Cockpit Voice Recorder
DFDR Digital Flight Data Recorder - a tape recording system that constantly records some aircraft flight instrument readings.
Ditching An emergency procedure by which an aircrew attempts to land a land-based aircraft on the water.
EMI Electromagnetic Interference
ESU Electrical System Upgrade. The C-130 programmed modification to provide no-break power to the sensitive, critical components.
FAA Federal Aviation Administration
FCF Functional check flight. The test flight done immediately following major aircraft maintenance.
FMEA Failure Modes and Effects Analysis
FMECA Failure Modes, Effects and Criticality Analysis
Fuselage tanks Fuel tanks in the cargo compartment of some C-130 aircraft.
Gravity Feed When fuel flows directly from a tank to its associated engine without the use of fuel tank boost pumps.
GTC Gas Turbine Compressor
IDCU Integrated Display Control Unit
INU Inertial Navigation Unit
ISO Isochronal inspection. A home station maintenance action requiring partial disassembly of the airframe.
LTM Lockheed Technical Manual
Master Engine Engine whose propeller RPM is used as the standard against which all other (slave) engines are synchronized.
NTSB National Transportation Safety Board
NVGs Night Vision Goggles
NVIS Night Vision Imaging System
OJT On-the-job training
PDM Programmed Depot Maintenance. Periodic major preventative maintenance performed on an aircraft at a maintenance depot.
Pogo Draining a small amount of fuel out of aircraft tanks to check for contamination.
PQDR Product Quality Deficiency Report
PSI (or psi) Pounds Per Square Inch
PSYOP (pronounced si-op) Psychological Warfare Operations
Rollback An event in which multiple engines experience a sudden, relatively small, and simultaneous reduction in engine RPM--uncommanded by the crew and with no prior indications of engine problems.
RPM Revolutions Per Minute. In reference to the C-130, 100% engine RPM is 13,820 RPM and 100% propeller RPM is 1020 RPM.
SCNS (pronounced "skins") The self-contained navigation system.
SECAF Secretary of the Air Force
Slave Engine Engine whose propeller blade angle is adjusted, via the synchrophaser, to remain in phase and pitch with the master engine.
SPO System Program Office. The Air Force office which handles life cycle management of a weapon system.
Sump Draining a tank at its lowest point to check for and drain water or any other contaminants.
Tank-to-Engine A fuel management configuration in which each engine is fed by the corresponding fuel tank above it.
T.O. Technical Order - Technical Manuals, which outline operations and maintenance of Air Force aircraft, vehicles, and equipment.
TCTO Time Compliance Technical Order - a Technical Order, which must be complied with in a certain period of time.
TD Temperature Datum System, which is composed primarily of the TD valve and TD amplifier (TD amp).
"NULL" Refers to the condition when the TD Amp sends no signal to the TD valve.
"PUT" Refers to the condition when the TD Amp sends a signal to the TD Valve to add fuel.
"TAKE" Refers to the condition when the TD Amp sends a signal to the TD Valve to restrict some fuel.
TIT Turbine Inlet Temperature
UHF Ultra High Frequency
USAF United States Air Force
VHF Very High Frequency

9.2 Team Members and Advisors

C-130 Broad Area Review Team Members

Name

Organization

1

Floyd, Maj Gen Bobby HQ AMC/LG

2

Bair, Mr James HQ AFMC/EN

3

Siegel, Col Gregory WR-ALC/LB

4

Snedeker, Col Mike HQ AMC/XP

5

Kane, Col Bill 908 AW (AFRC)

6

Sandiford, Col Gary AF/ILM

7

Macken, Lt Col Jerry AF/XOOX

8

Stanton, Lt Col Dan AFSC

9

Braden, Lt Col Bill HQ AMC/LGAA

10

Williams, Maj Roger AFSC (HC/MC-130)

11

Lesmerises, Capt Alan SA-ALC/LPEBT

12

Cannon, Capt Mike HQ AMC/LG EXC.

13

Wisener, CMSgt Randy HQ AMC/DOV

14

Love, SMSgt Donald M. 145th AW N.C. ANG

15

Nierescher, TSgt Kevin Patrick, (HC-130 Flight Eng)

16

Babcock, TSgt Gary 182AW IL, ANG

17

Anderson, Mr George NTSB

18

Eubanks, Mr Phil WR-ALC/LBR

19

Jones, Mr Rick WR-ALC/LBR

20

McGregor, Mr Ron AFSC

21

Puckett, Mr Ben WR-ALC /LBR

ADVISORS

Name

Organization

1

Chappell, Col David W. HQ AMC/JA

2

Leong, Lt Col Linda HQ AMC/PA

3

De Castro, Mr Al LMASC

4

Scheurich, Mr Scott Allison

5

Hack, Mr Bill Hamilton Standard

6

Maternowski, Mr George Hamilton Standard
Top Articles
Latest Posts
Article information

Author: Corie Satterfield

Last Updated: 02/05/2023

Views: 6645

Rating: 4.1 / 5 (42 voted)

Reviews: 81% of readers found this page helpful

Author information

Name: Corie Satterfield

Birthday: 1992-08-19

Address: 850 Benjamin Bridge, Dickinsonchester, CO 68572-0542

Phone: +26813599986666

Job: Sales Manager

Hobby: Table tennis, Soapmaking, Flower arranging, amateur radio, Rock climbing, scrapbook, Horseback riding

Introduction: My name is Corie Satterfield, I am a fancy, perfect, spotless, quaint, fantastic, funny, lucky person who loves writing and wants to share my knowledge and understanding with you.