NTSB challenges Boeing estimates of 787 battery safety
Assumptions used to certify the battery system proved wrong, the NTSB said Thursday.
Seattle Times aerospace reporter
In a blunt briefing Thursday, the chief of the National Transportation Safety Board (NTSB) said the government and Boeing badly underestimated the possibilities for failure of the 787’s lithium-ion batteries.
Boeing also got a bit of good news, as it won permission to begin test flights of the cutting-edge jet that’s been grounded for three weeks. But the sweep of the NTSB investigation, and the agency’s critique of the plane’s testing and certification, suggest the jet-maker has a long way to go before passenger flights can resume.
The NTSB chief said the safety assessments Boeing and the Federal Aviation Administration (FAA) made while certifying the 787 lithium-ion batteries for commercial service “were not borne out by what we saw in flight experience.”
Boeing estimated, for instance, that a battery problem would cause smoke less than once during 10 million hours of flight, while the plane so far has experienced two incidents in less than 100,000 hours.
Former safety-board member John Goglia said in an interview afterward the briefing that the statistical standard for safety of the 787 batteries had been set too low by the FAA — and “Boeing didn’t even meet the reduced standard.”
The NTSB briefing suggested that Boeing will have to redesign and recertify the jet’s battery system for a long-term fix.
“The assumptions used to certify the battery must be reconsidered,” said NTSB Chairwoman Deborah Hersman. “The design and certification assessment, and the assumptions that were made, were not borne out by what we saw in flight experience.”
The FAA grounded the plane Jan. 16 after a Jan. 7 battery fire at the Boston airport, followed just over a week later by a second event, when a battery smoldering on a flight in Japan forced an emergency landing.
Those incidents make clear, said Goglia, that “Boeing didn’t identify and mitigate all the failure modes.”
After the NTSB briefing, Boeing issued a statement that defended its 787 test and certification process as “rigorous” and “extensive,” while conceding that it may need improvement.
“We are working collaboratively to address questions about our testing and compliance with certification standards,” the statement said. “We will not hesitate to make changes that lead to improved testing processes and products.”
Flight testing of the 787 will resume within days.
The FAA on Thursday granted Boeing permission to conduct tightly controlled test flights so it can gather data and move forward with tests of potential fixes.
And Thursday morning, the first 787 to fly in the U.S. since the fleet was grounded arrived in Everett from Texas.
That wasn’t a flight test, but a “ferry flight” without passengers that the FAA allowed so Boeing could reposition the jet.
One in 10 million
Hersman said Thursday the investigation has pinpointed the start of the Japan Airlines battery fire as a short circuit inside a single cell of the eight-cell battery.
The overheating of cell No. 6 spread to all the others and caused the fire, she said.
Hersman said Boeing had studied the possibility of such a single-cell short circuit and its effects, in tests during the certification process.
Boeing concluded that these tests “showed no evidence of cell-to-cell propagation or fire in the battery,” she said.
However, in the fire on the Japan Airlines jet at Boston’s Logan airport, that’s exactly what happened.
In another certification test, Hersman said, Boeing studied the possibility that a failure in a single cell would result in smoke emission from the battery and estimated this would happen “less than once in every 10 million flight hours.”
“The 787 fleet has accumulated less than 100,000 flight hours,” she said.
“Yet there have now been two battery events resulting in smoke less than two weeks apart on two different aircraft.”
Goglia said the standard FAA safety requirement to cover the failure of any system critical to flight is that it must not occur more than once in a billion flight hours — which means such a failure is not expected to occur in the life of the fleet.
So the 787 battery-safety requirement that assumed a failure once every 10 million flight hours — 100 times more frequently — is already a low bar by FAA standards, he said.
Hersman said the NTSB will “examine the safety certification process used by both the FAA and Boeing for the 787 battery design and determine why the hazards identified in this investigation were not mitigated.”
Root cause sought
The NTSB still hasn’t identified the cause of the initial short circuit but has narrowed it down to possibilities.
One is some defect inside the battery due to the manufacturing process.
Each of the eight cells contains about 100 feet of metal foil wound and folded over itself with sheets of plastic separators between the layers.
These windings contain the electrodes, the negative made from graphite-coated copper, the positive from lithium cobalt-dioxide-coated aluminum.
“We are looking for evidence of contamination of electrode folds, wrinkles and pinches in the assembly of the cells and the battery,” said Hersman.
Another option is an inadequate battery design.
“We’re looking at the total design of the battery, including the physical separation of cells, their electrical interconnections and their thermal isolation from each other,” Hersman said.
The final possibility being investigated is some malfunction in the battery charging system, which could be in the controllers outside the battery or in the charge monitors within the battery.
Whichever of these is found to have started the overheating, the battery safety features failed to cope with the initial failure and the NTSB analysis makes clear Boeing will have to revisit the entire system design.
Hersman said the NTSB also is exploring the larger question of whether lithium-ion battery technology is well-developed and safe enough to use on airplanes at all.
“Lot of learning to do”
“We’re all trying to get smarter here at the NTSB on lithium-ion batteries,” she said. “There’s a lot of learning to do.”
Goglia said the cobalt-dioxide ion used in the 787 battery is “the most volatile chemistry” in lithium-battery designs. Other options — such as lithium-ion phosphate — are less volatile and could be safer.
Boeing “may be able to get away with changing the chemistry” rather than switching to a different battery type, Goglia said.
Hersman promised an interim factual report on the NTSB investigation within 30 days. But she doesn’t expect that report to be final or conclusive.
“Design, certification and manufacturing — these are all still on the table and we have a lot of work to do,” Hersman said.
She emphasized that the FAA is the agency that will decide when the 787 can carry passengers again.
Once Boeing develops a fix, it will have to certify the battery system all over again.
Given the pressure on the FAA from the NTSB findings, Goglia said he expected the regulatory agency to be very meticulous, “to insist that every T is crossed and every I is dotted.”
“It will take additional time,” Goglia said. “Months.”
In any case, it seems very unlikely the FAA will move to lift the grounding of the 787 in-service fleet before the interim NTSB report comes out.
Thursday afternoon, Boeing Commercial Airplanes CEO Ray Conner sent a message to the company’s union engineers, who are voting by mail on an option to strike over a new contract offer.
He urged the engineers to vote “no” on a strike and to come together as a Boeing team.
“We can do anything together, including emerging from our recent difficulties with the 787,” Conner wrote.
Dominic Gates: (206) 464-2963 or firstname.lastname@example.org