NTSB’s final reports delayed by agency’s limited resources

 - September 15, 2006, 11:32 AM

An aircraft accident happens, on average, three or more times a day in the U.S. If people die in such an accident, investigators from the NTSB are sure to be involved in trying to figure out what happened. And if the accident was prominent enough to be mentioned in the news, an NTSB spokesperson will probably say that the agency is not able to speculate about the cause and that a final determination of probable cause won’t be available until the NTSB report is completed.

The aviation industry is used to that procedure and expects NTSB final reports to appear, on the www. ntsb.gov Web site, not much longer than a year after the accident. But NTSB reports are not always issued within a one-year period; some are done sooner, and many take longer–much longer–than a year to finalize.

The NTSB and its investigators are a vital component of aviation safety. We all learn from each accident (and probably believe that reading about how others messed up somehow makes us immune). So when accident reports are delayed, for whatever reason, the aviation community and the flying public suffer. Safety information that isn’t available is of no use to anyone. And although NTSB investigators often release preliminary factual information that can be helpful to those who look for it, they are under no obligation to release such information before publication of the final report.

In some accident cases, NTSB investigators, who suffer from lack of resources (time and money), are way behind on delivery of final accident reports. Some of these delayed reports involve critical safety issues, and there is no good answer about why these reports are delayed, when they will be completed and what the NTSB is doing to try to ease the logjam.

One notable accident that involves important safety information is the crash of an MD 500 helicopter during a sightseeing flight in Volcanoes National Park near Hilo, Hawaii, on June 15, 2003. During the three years since the accident, the NTSB has yet to release anything more than the preliminary report.

That accident, caused by failure of the Rolls-Royce 250 engine, has generated three FAA Special Airworthiness Information Bulletins, one Airworthiness Directive and multiple lawsuits. However, NTSB investigator-in-charge Tealeye Cornejo has yet to publish any available factual information or the final report and probable cause of the accident. Meanwhile, operators of aircraft powered by the Rolls-Royce 250 are left in the dark about the safety implications of this accident, although at least they do have the FAA bulletins and AD.

In March AIN submitted a Freedom of Information Act request to the NTSB for any factual information available on this accident. While the Safety Board acknowledged AIN’s request within the required period of time, at press time it had yet to provide any of the requested material. Cornejo did respond to a request for an update last month. “The report is not complete,” she wrote in an e-mail, “and there is nothing new. The report should be released no later than the end of this year.”

To be fair to the NTSB, many of the preliminary and factual reports in any year of the database either have blindingly obvious crash causes (such as buzz jobs gone awry or flying into a box canyon, for example) or the crashes are mysteries that might never be solved. Typical mystery accidents include airplanes that ditch in the ocean and sink, killing all aboard, and the numerous non-U.S. accidents that could contribute to the body of aviation safety knowledge but whose investigations appear to be bogged down in bureaucratic inertia.

A Long Road to a Probable Cause Statement

A random look through the NTSB database revealed that there are plenty of old accidents, many of which were issued probable cause reports only recently, and many in which probable causes have not been determined and important safety
information might still be revealed. Some of the accidents are surprisingly old and beg the question of why final probable-cause determinations take so long. Examples include:

• Incident: Oct. 10, 2001. Probable cause report: May 30, 2006–A jammed stabilizer trim actuator on a Boeing 737 forced the crew to declare an emergency and land. The probable cause was “the failure of the stabilizer electric trim motor as a result of internal mechanical seizure.”

• Accident: Feb. 4, 2003. Probable cause report: May 30, 2006–An overheated inverter in a flight test Bombardier jet caused a fire that was found during post-flight inspection. The probable cause was: “The uninterrupted current overload of the cabin outlet system static inverter leading to ignition and sustained fire. The cause of the initiating overload condition could not be determined. Contributing factors were the lack of adequate current overload protection for the static inverter due to a failure to adequately assess fire/smoke hazards of the inverter in this installation and the absence of timely fire annunciation to the flight crew.” The inverter was redesigned.

• Accident: May 31, 2001. Probable cause report: April 25, 2006–A propeller bolt failed in an agricultural Schweizer G-164B.

• Accident: Dec. 25, 2001. Probable cause report: April 25, 2006–The pilot ditched into the ocean after magneto failure caused the engine to stop. Despite the fact that mechanics had complied with AD96-12-07 on the magneto’s impulse coupling, the coupling’s spring failed, causing altered magneto timing and subsequent engine failure.

• Incident: Oct. 5, 2003. Probable cause report: April 25, 2006–One engine on an Airbus A330 suffered an uncontained failure on climbout. Carbon deposits believed to have accumulated in an engine vent tube caught fire, burning through vital components, which led to the failure. The Airbus was departing on an ETOPS flight and returned safely to Miami International Airport.

“Contributing to the cause of the uncontained engine failure,” wrote the NTSB in the probable cause statement, “was the absence of measures to adequately monitor the in-service performance of a new engine/oil combination.”

• Accident: May 2, 2002. Probable cause report: not yet issued. A NetJets Cessna Citation Ultra ran off the runway while landing at Real County Airport near Leakey, Texas. No one was injured but the airplane was destroyed in a post-impact fire. A witness reported seeing the Ultra touch down halfway down the 3,975-foot runway. This was later confirmed when investigators observed tire marks that were 2,100 feet from the landing threshold. The calculated total stopping distance, according to the NTSB’s report on flight manual numbers, was 2,955 feet. While nearing the runway, according to the NTSB factual report, “the PIC noticed a 16-knot increase in speed above reference.”

Many accidents generate all sorts of FAA and NTSB activity, ranging from NTSB safety recommendations to FAA advisories and Airworthiness Directives. One important link in the safety chain missing from the NTSB database is any mention of the safety efforts the accident precipitated. It could be tremendously helpful for the aviation industry to be able to read about a particular accident then be able to follow links to the safety improvements associated with each accident.

For example, the FAA’s recent issuance of a new policy mandating a 15-percent safety margin for Part 91K operations like that of the fractional Ultra above might be in part a result of that accident. A more recent accident, that of a Southwest Airlines 737 that overran a contaminated runway at Chicago Midway Airport, is cited as the reason for the policy change, but the Ultra accident might also be related.

Like many government agencies, the NTSB is strapped for resources and must prioritize its activities. In some of the above accidents, it seems likely that had there been fatalities, the Safety Board would have paid more attention to the causes and the final reports would probably have been turned around sooner.