Twelve Years of Half Measures

Wed, Aug 27, 2008 — David Evans


A Mute Testimonial to NTSB Impotence

While Canada’s Transportation Safety Board (TSB) has achieved only about a 20% acceptance rate for its Swissair flight 111 recommendations, the U.S. National Transportation Safety Board (NTSB) may not have fared much better in its TWA flight 800 recommendations, even though a general comparison indicated a 40% acceptance rate.

The TSB’s accounting is generous, and even the one-in-five acceptance rate that it has attained in the ten years since the Swissair disaster in Halifax may be a stretch (see Aviation Accident & Incident Digest, ‘A Decade of Minimal Action,’ homepage).

The NTSB experience following the TWA flight 800 disaster may be similarly frustrating. According to the NTSB’s accounting of recommendations accepted, those that have been rejected or not acted upon by the Federal Aviation Administration (FAA), and those recommendations still open because of promised action, the Board has attained a two-in-five implementation rate. Just six of 15 recommendations have been closed out some twelve years after the accident with an “Acceptable Action” rating.

A close reading of the voluminous correspondence concerning the recommendations reveals a similar FAA pattern of delay, partial measures, and outright obstructionism.  Part of the problem lies with the NTSB recommendations, which strike one as duplicative, of secondary or tertiary importance, or avoiding altogether the big issues – such as why fuel tank inerting has not been required all along, and why flawed electrical wiring insulation (which also afflicted the Swissair jetliner) has not been subject to stringent risk analysis and maintenance protocols.

Fuel tank inerting will finally be required for some aircraft (those not likely to be retired in the next few years), and inerting has only been mandated for in-fuselage center wing tanks with heat sources nearby – although the NTSB urged inerting of all fuel tanks. The FAA has yet to call for insulation between those fuel tanks and the equipment generating the heat. Moreover, the FAA has not prohibited the placement of heat-generating equipment in close proximity to fuel tanks, thus perpetuating an undesirable design for years to come.

Similarly, with wiring, the FAA has yet to establish in regulation a prohibition against the running of high-power circuits in the same bundle that contains signal wiring, nor has the FAA established separation criteria for aircraft wiring. As a result, many of the heat-related and electrical problems found in both the Swissair and the TWA investigations remain uncorrected and could be designed into the next generation of airliners.

Wires Affecting Multiple Systems Routed in the Same Bundles


Schematic diagram of the B747-100’s FQIS wiring, fuel flow wiring, cockpit voice wiring and other wiring information. In the twelve years since the TWA flight 800 accident, the design practices remain unchanged and approved by the FAA.

Diagram: NTSB, TWA flight 800 accident report, AAR-00/03

When all is said and done, the NTSB’s success rate with the TWA flight 800-related recommendations seems closer to the 20% attained by the TSB regarding the Swissair flight 111 recommendations. Below, a detailed accounting of the responses to the TWA flight 800-related recommendations, in which promised action, without timely follow-up, seems enough to satisfy the NTSB. Passengers, if they took the trouble to find out, would be dismayed – if not horrified – at the clear pattern of inaction. The color-coding – green, yellow, red – is an artifice of this publication, not the NTSB, to highlight the status of each recommendation; as the comments imply, many recommendations color-coded yellow could just as easily have been coded red:

Recommendations issued in 1996

– Recommendation A-96-174

That the FAA develop and implement changes that will preclude the operation of airliners with explosive fuel-air mixtures in the fuel tank, including inerting systems and the installation of insulation between heat-generating equipment and fuel tanks. The modifications should apply to newly certificated airplanes and, where feasible, to existing airplanes.

Rating: Open – Acceptable Response

(Comment: The FAA announced 21 July 2008 that new airplanes with heated center wing tanks must be equipped with inerting within two years, and aircraft already in the fleet must be retrofitted with inerting systems within ten years [see Aviation Safety & Security Digest, ‘Significant Regulatory & Related Activity’]. Thus, a window of some 22 years from the TWA flight 800 accident until the last airplane is retrofitted is allowed. The ruling makes no mention of the NTSB’s call for insulation between equipment that generates heat, such as air conditioning packs, and the fuel tank. Moreover, the FAA only requires inerting on heated center wing tanks, while the NTSB called for inerting of all tanks, center and wing, heated and unheated. It should be mentioned that the FAA propounds a substantially lower safety standard for airplane fuel tanks than those implemented on ground based flammable liquid storage tanks, which, by the way, do not feature people seated in close proximity. Reflect also on the likelihood that the reason insulation above the underlying air conditioning packs was not mandated is itself rooted in a built-in design flaw. The fuel tanks were intended as a heat-sink of the high temperatures produced on the ground by the air conditioning units immediately below. If it wasn’t for the overlying center-section tank, the same level of air conditioning cooling could only have been achieved through large diameter air-ductwork incorporating high capacity throughput fans – constituting what was perceived as an intolerable weight penalty as well as a volumetric capacity problem [i.e., the size and shape of the center-tank would have been restricted]. Whether the resulting generation of flammable fuel/air fumes in the ullage was ever considered is unknown. However, it is undeniable that this easy-out design facet was a root cause of TWA flight 800’s demise, as the packs were operated on the ground for at least two hours before takeoff.)

– Recommendation A-96-175

That the FAA develop procedures, pending implementation of design modifications, to reduce the potential for fuel tank explosions by operating with a minimum fuel quantity, monitoring the fuel tank temperature, or refueling from cooler ground fuel tanks.

Rating: Closed – Unacceptable Action

(Comment: The FAA indicated it is not considering or encouraging any short-term measures to address this recommendation. Even with the inerting system announced above, the FAA will not require cockpit monitoring of fuel tank temperature or the percentage of oxygen concentration which, if implemented, would provide unambiguous indication to the flight crew that the tanks are inerted.)

– Recommendation A-96-176

That flight crews be informed of the center wing fuel tank temperatures found by flight tests, and be informed of operational procedures to reduce the potential for exceeding center wing tank temperature limitations.

Rating: Closed – No Longer Applicable

(Comment: The FAA claimed that TWA no longer operated the B747-100, the flight 800 accident aircraft, and therefore this recommendation was superseded by events. However, the recommendation called for action “on other aircraft.” For example, the B737, B757, B767 and B777 feature heated center wing tanks. It would seem that a more appropriate characterization of this recommendation would be “Closed – Unacceptable Response.”)

– Recommendation A-96-177

That the FAA require modification of B747 and other airplanes with fuel tanks near heat sources to incorporate cockpit displays of fuel tank temperatures.

Rating: Closed – Unacceptable Action

(Comment: The FAA claimed that flights would be canceled if the ground temperature were above 30˚ F, as this limitation would be necessary to avoid entering the flammable regime at the top of climb. The NTSB said it is “disappointed that the FAA does not believe in the safety benefit of this recommendation.”)

Recommendation issued in 1997

– Recommendation A-97-011

That the FAA develop a checklist to be used by explosives detection teams to prevent contamination of aircraft. This recommendation was based on the following, according to the NTSB: “A dog handler, working under the auspices of the FAA’s K-9 Explosive Detection Team Program, had spilled trace amounts of explosives while placing training aids on board the (accident) aircraft during a proficiency training exercise.”

Rating: Closed – Acceptable Action

(Comment: The FAA said it was using barrier materials to prevent leaving explosive residues on surface areas, and it developed a training log in lieu of creating an additional checklist.)

Recommendations issued in 1998

– Recommendation A-98-034

That the FAA issue, as soon as possible, an airworthiness directive (AD) to inspect B747 classic fuel quantity indication systems (FQIS) for possible damage.

Rating: Closed – Acceptable Action

(Comment: The FAA issued AD 99-08-02 on 29 March 1999, requiring such inspections and replacement of probes with sharp edges that could damage FQIS wiring. Note that the AD was not issued until the following year, despite the NTSB’s call for inspections “as soon as possible.”)

– Recommendation A-98-035

That the FAA issue an AD requiring earliest possible replacement of Honeywell series1-3 terminal blocks on B747 fuel probes.

Rating: Closed – Acceptable Action

(Comment: See above. Note that “bridging” [i.e., arcing across] of conductive sulfide deposits between positive and negative terminals is thought to have led to TWA flight 800’s fatal spark – once higher currents and voltages had been induced into the FQIS wiring via faults elsewhere in the aging aircraft’s wiring bundles.)

– Recommendation A-98-036

That the FAA survey other aircraft-model FQIS probes to determine whether potential fuel tank ignition sources exist that are similar to those found on the B747, and repair or replace any damaged wiring.

Rating: Open – Acceptable Response

(Comment: The NTSB’s status is outdated, as it hasn’t been updated since 2000. The FAA noted that removal of wires in response to this safety recommendation could lead to errors – in other words that the ”cure” could be worse than the “disease.”)

– Recommendation A-98-037

That the FAA require research into copper sulfide deposits on FQIS parts in fuel tanks to determine the levels of deposits that may be hazardous, and to clean or replace contaminated components.

Rating: Closed – Acceptable Action/Superseded


B747 FQIS probe showing evidence of sulfide contamination


B747 FQIS wiring showing sulfide buildup due to leaching through wiring insulation

(Comment: The NTSB said it was pleased that the FAA was continuing to support research into sulfide deposits and said that action should be taken to eliminate the risk of ignition posed by silver-sulfide deposits inside fuel tanks. How this recommendation was closed with a characterization of “Acceptable Action” is a mystery, as more work was clearly required. It’s no intellectual leap to see that replacing all existing silver-plated connectors and components with nickel-plated equivalents would be an almost impossible task – as would removing all sulfur from fuel. It was always a far easier proposition to meekly accept that conductive bridging deposits would accumulate over time, and to eliminate the other side of that lethal equation by eliminating all stray electrical voltages in fuel tanks by the strict inspection regimes mandated by Special Federal Aviation Regulation 88 [SFAR-88]).

– Recommendation A-98-038

That the FAA require FQIS wires routed in bundles with powered circuits be separated or shielded.

Rating: Open – Acceptable Response

(Comment: An NTSB scientist, who used to work on Air Force missiles, said the co-routing of signal wires and power circuits in the same bundle on a commercial airliner was in sharp contrast to Air Force practice on unmanned systems, where power circuits and signal wires were never entwined in the same bundle. There are four pages of dense text recounting the exchange of letters between the NTSB and the FAA in which the FAA says transient suppression devices [TSDs] or separation and shielding will protect against the introduction of excess energy into wiring inside fuel tanks. The issue of routing high-power and signal circuits in the same bundle was not addressed. Nor was the advisability of running wires and other electrical components inside fuel tanks in the first place. It is known that Qantas airline installed TSU’s on its B747 classic fleet; however, ensuing FQIS indication problems proved insolvable and the installation was determined to be a failure – largely due to the silver sulfide deposits.)

– Recommendation A-98-039

That the FAA require surge protection systems to prevent spikes in electrical power from entering FQIS wiring.

Rating: Open – Acceptable Response

(Comment: This recommendation gets to the heart of why the TWA jet exploded, as a surge of current outside the center wing fuel tank jumped to the FQIS signal wire, traveled into the tank, and there arced across a terminal block’s silver sulfide deposits, causing the dense fuel/air vapor to ignite. There are five pages of text recounting the FAA’s response to this recommendation. A couple of key passages are especially relevant. First, “Although airplane manufacturers generally provide protection for certain electrical circuits, there is no FAA regulation that specifies wire separation criteria or identifies which circuits must be protected …” Note the use of the weasel words “generally” and “certain,” which lack specificity and can cover a host of sins. The TWA flight 800 investigation reported, “The Safety Board’s review of FAA regulations revealed that the FARs [Federal Aviation Regulations] do not contain specific guidance regarding separation of electrical circuits/wiring.” There are still not adequate wire separation distances in the regulations, some 12 years after the accident. Second, “(It) will be assumed that for a TSD (Transient Suppression Device) to receive approval, it would have to be shown that the possibility of any such failure is extremely improbably [e.g., one in a billion] … If flammability in existing and future transport-category airplanes were eliminated or reduced … the Safety Board would agree with the FAA that either TSDs or separation and shielding … will provide adequate protection … Therefore, pending the timely completion of the FAA’s actions to eliminate or reduce flammability” the recommendation stands. In other words, inerting of the fuel tanks will cover the gaps in protection provided by TSDs or separation and shielding. As far as the “timely” provision to inert, the FAA published a final rule only recently, on 21 July 2008, requiring inerting of heated center wing tanks only, and specifying inerting for existing aircraft is to be accomplished over a ten year period. In brief, the FAA has not honored the intent of the NTSB’s original recommendation, which was to inert all tanks, not just those with heat sources near them [see Aviation Safety & Security Digest, ‘Significant Regulatory & Related Activity,’ home page].)

Recommendations issued in 2000

– Recommendation A-00-105

That the FAA examine manufacturers’ design practices regarding electrical bonding of components inside fuel tanks and change the standards, as necessary, to eliminate potential ignition hazards.

Rating: Open – Acceptable Response

(Comment: Uniform bonding standards among manufacturers have not been adopted and any attempt at harmonization of practices internationally is on hold. The recommendation is in an open status pending the NTSB’s review of the uniform standards. Some 12 years after the accident, it seems that another 12 years could pass before design standards are available for NTSB review. Classification as “Unacceptable” seems wholly appropriate.)

– Recommendation A-00-106

That the FAA review designs for aircraft wiring systems to (1) identify safety-critical systems and (2) require adequate separation for these systems.

Rating: Open – Acceptable Response

(Comment: The NTSB notes the following – “[A]lthough the FAA has initiated numerous actions, most of the efforts to date provide only guidance, rather than the needed changes to the FAA’s requirements. The Safety Board is also concerned that some of the FAA’s proposed regulatory revisions are inadequate. For example, the Board has learned that the November 2007 proposed aircraft wiring systems inspection requirements will only mandate a visual inspection, despite FAA-sponsored studies that have shown that visual inspections may miss half (or more) of wiring defects.” Actually, the situation is worse than that described by the NTSB. Wiring inspections have been mandated for the fleet, but these visual inspections will be limited to those areas within two inches of critical flight controls and/or those areas where lint has accumulated. It is not clear if even a quarter of the wiring in a modern jetliner will be subject to these admittedly inadequate inspections, which will not identify a failed wire buried inside a bundle – which requires a nondestructive test [NTD] to detect (see below.)

Wiring Inspection Perspective

Bundle in free space, and why visual inspection is inadequate


Maximum percent visible:         18.9%

Minimum percent inaccessible: 81.1%

Source: NASA

More than token inspections, this recommendation called for design and modification of safety-critical wiring, which has not been done. Classification as “Unacceptable” seems not only more suitable but necessary.)

– Recommendation A-00-107

That the FAA require corrective actions to eliminate the ignition risk posed by silver-sulfide deposits on FQIS components inside fuel tanks.

Rating: Open – Acceptable Response

(Comment: The recommendation seems redundant to A-98-037 above. However, as pointed out above, eliminating sulfur from fuel and silver from installed electrical components proved to be expensively impractical. Rather than elucidate these impracticalities, “other” solutions were imposed, notably SFAR-88.)

– Recommendation A-00-108

Within 90 days, the FAA should brief the Safety Board on the status of its efforts to address all of the issues in the aging transport non-structural systems plan.

Rating: Open – Acceptable Response

(Comment: Six pages of dense text outline the “letter war” between the NTSB and the FAA on this recommendation. The 90 days for a satisfactory and complete response has clearly passed and characterizing this recommendation as “Unacceptable” seems more appropriate. The NTSB noted that the FAA’s Service Difficulty Report (SDR) system has not yet been rebuilt to support trend analysis. What the NTSB doesn’t say is that SDRs are only submitted when problems occur in flight, and that ground glitches are not even reported. Moreover, reporting discipline among airlines varies greatly, and that some operators don’t submit SDRs at all. Regarding arc fault circuit breakers (AFCBs), which are regarded as a great advance in protection against the ravages of electrical arcing, the Board said, “Although the FAA cites the implementation of AFCB technology, this implementation will be limited. The Board also notes that 5½ years after the FAA identified aging mechanical systems as an issue … the FAA has not yet completed research, let alone translated that research into changes in maintenance.” Delay triumphs, and it’s especially ironic in this particular recommendation, with its somewhat sophomoric call for positive action within 90 days.)

Comments are closed.

Nolan Law Group