Tests of Boeing Co.'s 787 Dreamliner batteries weren't adequate to ensure that they wouldn't overheat or burn, the top U.S. aircraft-accident investigator said after a two-day hearing.
Officials from Boeing and the U.S. Federal Aviation Administration, which approved the batteries, also said at the National Transportation Safety Board hearings that concluded yesterday that the tests didn't anticipate the two failures that led to the grounding of the company's newest airliner.
"The testing was not as conservative as it could have been," NTSB Chairman Debbie Hersman said yesterday in a news briefing.
Hersman, in some of the most detailed comments since the agency began probing a Jan. 7 fire aboard a 787 on the ground in Boston, said the NTSB hoped to determine a cause for the battery failure within a year. The safety board pressed Boeing, its subcontractors and the FAA during the hearing on how they certified the batteries.
The Dreamliner has been grounded since a second battery incident Jan. 16 in Japan, the longest U.S. grounding of a commercial plane in the jet age.
The 787 will return to service in a matter of weeks or sooner, after U.S. regulators on April 19 approved battery protections to contain a fire and keep fumes from reaching the cabin. The European Aviation Safety Agency approved the fixes April 23 for carriers in that region, according to a statement.
The failure on a Japan Airlines 787 Jan. 7, which spread to all eight of the battery's cells after a short circuit in one, violated the FAA's safety conditions for the Dreamliner, set in 2007, Hersman said. One prohibited a cell failure from triggering uncontrolled temperature increases that would spread to other cells, a condition known as a "thermal runaway."
"That's exactly what we saw in Boston," Hersman said.
The FAA's engineers assigned to oversee the 787 knew little about lithium-ion battery technology and had to learn on the job while assessing Boeing's design, an agency manager said.
The engineers were new to the subject when Boeing proposed using lithium-ion batteries on the model, Steve Boyd, manager of the FAA's airplane and flight crew interface branch, said today.
"When the first lithium-ion proposal was made, we had people who were expert in batteries, but not necessarily specifically with lithium-ion technology," Boyd said.
While FAA engineers may not be considered lithium-ion specialists, they have learned since those early days and are capable of assessing the technology's safety, Boyd said.
"We had an extremely experienced group of people working on this," he said. The average experience of the agency's team of engineers on the project is about 30 years, he said.
Hersman, after saying on April 23 that a top Boeing executive had engaged in "obvious obfuscation," yesterday turned her attention to why the FAA signed off on the batteries.
She asked whether the batteries had met the safety standards placed on them by the agency.
"I don't necessarily know how to answer that particular question," said Jerry Hulm, a Boeing systems engineer.
The 787 received two to three times as many safety tests as Boeing's previous model, the 777, said Doug Lane, Boeing's director of commercial-airplane regulation. That was because of "increased rigor" by the FAA, Lane said.
The FAA defended its use of Boeing employees to sign off on some new aircraft approvals. Delegating such authority to employees of aircraft manufacturers has been done since the 1920s, said Dorenda Baker, the agency's aircraft-certification service chief.
"Delegation, I want to emphasize, does not mean lack of FAA involvement," Baker said. The agency had final authority on decisions by Boeing employees, she said.
Company employees were allowed to approve designs they had worked on, Lane said in response to a question.
Boeing and the aviation agency were pressed by safety board officials on their assumptions that there was little chance a battery could fail, and on testing methods that led to that conclusion.
"We used the state of the art in the industry at the time," Mike Sinnett, a Boeing vice president and the 787's chief engineer, told the safety board.
Boeing concluded in tests before the plane was certified to fly that an internal short circuit in the battery system couldn't trigger a fire, according to preliminary reports from the board. The company shot a nail into a battery cell to simulate a short and it didn't burn.
Boeing has since concluded that the nail test isn't a good way to simulate an internal short, Sinnett said. "We don't feel that it was conservative enough."
Without knowing what triggered the two battery incidents, it's impossible to know whether the FAA's conditions for battery safety were met, Boyd said.
A 2009 battery failure at a company that made components for the electrical system was caused by an internal short- circuit. Details of the event at a Hamilton Sundstrand Corp. plant in Rockford, Illinois, which hadn't been disclosed, appeared in records released April 23. The safety board says multiple shorts triggered the 787 fire in Boston on Jan. 7.
GS Yuasa Corp. of Kyoto, Japan, made the 787 battery packs. The charger is manufactured by Tucson, Arizona-based Securaplane Technologies Inc. Both suppliers sell the products to Thales SA, which then provides them to Boeing.
GS Yuasa redesigned the battery twice to improve safety before the 787 entered service, including after the 2009 incident in Rockford, Takahiro Shizuki, manager of GS Yuasa's large lithium-ion battery technical unit, told the safety board.
The agency is still trying to identify the underlying causes of the Boston fire. The reasons a battery aboard an All Nippon Airways Co. overheated and released fumes in Japan during the second incident on Jan. 16 also haven't been determined. The airline changed its name to ANA Holdings Inc. this month.
Boeing, which has an order backlog for more than 800 Dreamliners, with a list price starting at about $207 million, halted deliveries during the grounding.