Unfriendly Skies

05.28.14

NTSB Doesn’t Think the Boeing 787 Dreamliner Is Safe Enough to Fly

The agency that investigates every airplane accident is telling the FAA to look again at hazardous lithium-ion batteries after the entire fleet was cleared for takeoff following two accidents last year.

There are 140 Boeing 787 Dreamliners now flying with airlines across the world. Passengers love the 787’s airy, spacious cabin, its large windows and its cutting-edge entertainment technology. Last year’s problems with battery fires that caused the entire fleet of 787s to be grounded for three months seem well behind it. Or are they?

The National Transportation Safety Board is at odds with the Federal Aviation Administration over the way the 787 was cleared to fly again. The Board has raised new concerns about the way that the lithium-ion batteries used to supply power to critical systems on the 787 are tested to ensure their safety—even suggesting that more rigorous testing is needed.

In fact, lingering questions and confusion about the danger posed by the batteries point to a lack of transparency in the way both Boeing and the regulators deal with the application of new and relatively untried technologies in airplanes. The NTSB is plainly unhappy with the steps taken so far by the FAA to understand the specific risks posed by lithium-ion batteries: “Aircraft manufacturers,” says the Board, “need to evaluate whether additional requirements and testing are necessary to ensure aircraft-level safety.”

The NTSB’s investigation into the first of the two emergencies that caused the grounding of the 787, a fire in a Japan Airlines 787 while it was empty and parked at a gate at Logan International Airport in Boston on January 7 last year, is still open, even though late in April 2013 the FAA accepted Boeing’s changes to the installation of the battery system as sufficient to get the fleet back in the air.

Thousands of hours of testing by the NTSB, Boeing, the battery manufacturers and various other technical teams have failed to exactly replicate what happened to cause the Boston fire. It has been established that a short circuit caused what is called a thermal runaway in one cell of a battery which then spread to other cells and ended up frying the whole unit—and causing extensive damage to the rear electronics bay where the batteries were housed.

What eludes investigators is what triggered the short circuit.

The NTSB does not regulate, it investigates and makes recommendations to the FAA as a result of what it discovers during an investigation.

In grounding the 787, the FAA declared that the fires represented an “unsafe condition.” It was then up to Boeing to persuade regulators that it could fix the problem in a way that made the 787 safe to return to airline service. Without being able to find an initial trigger for the thermal runaway, Boeing assured the FAA that if the same thing happened again the airplane would not be endangered. They had sealed the batteries in an armored box which, they said, ensured the absence of oxygen to fuel a fire and, at the same time, they installed a pipe designed to vent gases from the box and out of the airplane.

Ray Conner, the president of Boeing Commercial Airplanes, said that this fix would “keep any level of battery overheating from affecting the airplane or being noticed by passengers.”

150527_cliver_batteries_787-embed
Jonathan Ernst/Reuters

It was on this assurance that the FAA certified that the 787 was safe to fly again.

The NTSB does not regulate, it investigates and makes recommendations to the FAA as a result of what it discovers during an investigation. Last week it made five recommendations directed at the testing and certification of lithium-ion batteries.

The last of these recommended that “the FAA develop a policy to establish…a panel of independent technical experts to advise on methods of compliance and best practices for certifying the safety of new technology to be used on new or existing aircraft.”

On the face of it, this was odd because such a body already exists, as the NTSB must have known. Indeed, the FAA immediately swatted away this recommendation, saying that it was “already working closely with domestic and international battery experts…to incorporate information learned from the experience with the Boeing 787.…”

That body is a Special Committee of an organization based in Washington with a quaintly dated name, the Radio Technical Commission for Aeronautics, more usually known simply by its acronym, RTCA.

The Special Committee was established in 2011. Its specific mission is to update the standards used for testing lithium-ion batteries so that they can be certified as safe for use in airplanes. When those deliberations are complete those standards will then be adopted by the FAA when it certifies that an airplane is safe to fly—in the case of the 787 these standards will be applied retroactively.

There are two problems with this arrangement.

The first is that deciding what the safety standards should be has become the pursuit of an ever-moving target. The more that lithium-ion battery technology develops, the more it appears that the battery industry doesn’t fully understand the ways in which the technology can fail. A commonly-used euphemism for this phenomenon is that the technology “is not yet mature.”

The second is the status of the RTCA. It calls itself a public-private partnership, but its authority is fudged under the label of a “Federal advisory committee.”

Last year, when the then-head of the NTSB, Deborah Hersman, held a two-day public hearing to track the history of lithium-ion battery development and how it had been chosen for and used on the 787, she was openly frustrated by the lack of information supplied by the RTCA and its Special Committee—she pointed out that its deliberations were not publicly accessible and said “we must have access, not barriers” in the pursuit of knowledge about new technology.

The absence of transparency seems to be related to the composition of the membership of the Special Committee. It includes representatives from developers and manufacturers of lithium-ion batteries who are—as was made amply evident in last year’s NTSB hearing—lawyered up and ready to invoke proprietary secrecy and intellectual property rights to protect themselves from competitors. This materially affects what can be revealed about the behavior of batteries during testing.

I have also discovered that the chairman of the committee is a Boeing executive, Richard P Nguyen, and that there are two other Boeing representatives among the 32 members—and that, on occasion, the committee has held meetings at a Boeing facility in Virginia.

Nobody questions the need for Boeing to be a party to the process of establishing the safety standards. They have their own expertise to bring to bear. But their technical record during the initial certification of the 787 is not reassuring.

To begin with, Boeing was two stages removed from the design, development, testing and manufacturing of the batteries. The responsibility for delivering the 787’s novel electrical systems was sub-contracted to the French aerospace company Thales who, in turn, contracted a Japanese company, GS-Yuasa, to develop lithium-ion batteries that would be called upon to supply far more power than any batteries previously used in an airliner.

In 2007 the installation of lithium-ion batteries in two separate electronics bays of the 787 was certified as safe in a deal made between the FAA and Boeing. Last week the NTSB commented that the “FAA certification staff members relied primarily on Boeing’s expertise and knowledge to define the necessary tests and analyses for certification.” It added: “If the FAA had reached out to other experts…[they] could have recognized that the 787 methods of compliance were insufficient to appropriately evaluate the risks….”

As part of that deal, the FAA accepted Boeing’s contention that the likelihood of a thermal runaway was “extremely remote”—only once in 10 million flight hours. As it turned out, the two incidents last year occurred when the fleet of 50 Dreamliners then flying had logged only 52,000 flight hours.

The grounding cost Boeing hundreds of millions of dollars, much of it in undisclosed compensation to airlines as well as the cost, estimated at $465,000 per airplane, of the new battery installations. The regulators were under great pressure to accept the armored box fix.

In the interests of safety, the whole process of certifying the airworthiness of airplanes using new, first-of-a-kind technology should be opened up to public scrutiny. The hermetic nature and instincts of the RTCA leave the implication of a cozy, unaccountable arrangement between Boeing and the lithium-ion battery manufacturers over what can be disclosed, together with what seems to be a glacial pace of response to a serious challenge.

The NTSB’s investigation into what happened to trigger the short circuit in the batteries aboard the 787 parked at the gate at Logan is still open. The Board is also assisting the Japan Transport Safety Board in their investigation into the second battery emergency on January 16 of last year. In this case the consequences could have been disastrous, since the battery failure occurred during a flight and left the FAA no choice but to ground all 787s.

A Nippon Airways 787, flying from Yamaguchi to Tokyo, had been in the air for 18 minutes when the pilots were alerted to smoke and heat in the forward electronics bay, directly beneath them. They diverted to the nearest airport, Takamatsu, and were on the ground within 16 minutes. The 129 passengers were evacuated on the runway via emergency slides.

Photographs from the ANA 787 exhibited by the Japanese Safety Board revealed a battery reduced to a mass of blackened wires inside a badly distorted casing. Hot chemicals—vaporized electrolyte—had left a trail of toxic residue in the electronics bay. The location of this meltdown bothered Hersman at the time. She pointed out that this battery was “the final power source, should all other electrical generation fail.”

I am told that the Japanese investigators have been particularly rigorous in their examination of the battery manufacturer, GS-Yuasa. In the absence of being able to reproduce the cause of the short circuits that triggered both emergencies, two red flags have emerged from the manufacturing process: quality control and testing.

In a 2006 test, GS-Yuasa created the effects of a short circuit by driving a steel nail through a cell case, but there was no thermal runaway, and consequently Boeing accepted that a short circuit would have limited effects and not jeopardize the safety of the airplane.

However, in its new recommendations to the FAA, the NTSB stressed that this nail test on a pre-production battery did not in any way represent how the batteries were installed in the 787 or the realities of every day operations. And they pointedly quoted a laboratory report: “an internal short hazard is one of the most difficult to reproduce, yet it is the most important to solve to improve safety.”

The Japanese investigators are also being closely watched by the Airline Pilots Association of Japan. The director of the Association, Toshikazu Nagasawa, told the New York Times last year that he was unhappy with Boeing’s armored box solution that got the 787s back in the sky. “Boeing says that any battery fire will now go out on its own, so there’s no safety issue,” he said. “But that’s on paper. No pilot would ever want to keep flying with a fire on board, whether it’s in a metal box or not.”

To be sure, there have been no more battery meltdowns in a 787 since the ANA emergency landing. The metal box solution therefore remains untested. In the meantime the NTSB has now made it abundantly clear that it wants the FAA to do more to show why it is so confident that the 787 remains safe to fly.

“The NTSB recommends that the FAA review the methods of compliance used to certify permanently installed, rechargeable lithium-ion batteries on in-service aircraft and require additional testing, if needed, to ensure that the battery design and installation adequately protects against all adverse effects of a cell thermal runaway.”