A top Boeing official said today at a fact-finding hearing that a nail penetration test used to certificate the lithium ion batteries on the 787 in 2011 was later realized to be insufficient, one of several discoveries for the company and safety investigators over the course of a grueling, three-month grounding.
"What we've since found out is that the nail penetration test, while it was believed at the time to be representative of a failure inside the cell and it was state of the art at the time, in retrospect I believe we don't feel it was conservative enough," says Boeing vice president and 787 chief project engineer Mike Sinnett.
Sinnett's comments on the first day of a two-day hearing by the US National Transportation Safety Board (NTSB) illustrate how the two battery failures that caused a three-month grounding of the 787 fleet have changed how the company and the Federal Aviation Administration (FAA) approach lithium ion-driven power devices and, not least, the introduction of new technologies in commercial aircraft overall.
On 19 February, the FAA cleared the 787 fleet to enter service as soon as Boeing installs redesigned lithium ion battery kits - even though American and Japanese investigators have not yet found a root cause for the two "serious" battery failures in January.
In the absence of a clear reason for the battery failures, the success of the 787's return to service will be based on how well Boeing and the FAA have learned to anticipate and prevent unknown and isolated failures from causing larger problems on the aircraft.
The nail penetration test conducted prior to certification was supposed to accurately measure how the GS Yuasa-supplied lithium ion batteries reacted to a short-circuit inside a single cell of the eight-cell battery. Boeing wanted to know if over-heating by one cell would cause a chain reaction in other cells, an effect known as "thermal runaway" within a battery.
Looking back on the test now, Sinnett says, puncturing the cell with the nail produced a short circuit that was less energetic than what the batteries that failed in service actually experienced.
There was another, more rigorous, test that Boeing could have performed prior to certification, Sinnett says, but the company decided not to do it.
The other test would have inserted a heating element inside the cell. Such a method "adds significant energy to the cell in the process of making it fail, and as a result it can mask the results by being too conservative", Sinnett adds. "So we didn't want to use the test at the time because we felt it wasn't representative of a failure inside the cell."
Left unsaid during the hearing, which was focused solely on the batteries that failed, was that Boeing used the heating element method to test the thermal runaway effect on the redesigned battery, and it passed the test.
In the first few hours of the NTSB hearing, it was already clear that the two-day series of testimony by witnesses will shed little light on the elusive root cause of the battery failures, which occurred on two aircraft less than 10 days apart in January.
According to Sinnett, Boeing has found no evidence of overcharging, which he identified as the most likely cause of a thermal runaway event, in either of the battery incidents. Neither is there any evidence of any other design or "building" error by the supplier. Sinnett has said previously that a root cause for the battery failures may never be known.
Another topic addressed in the hearing was how the FAA handles new technology during the certification process. The 787 entered service six years after Boeing first informed the FAA that the aircraft would use lithium-ion batteries. The FAA set the special conditions for Boeing to prove the airworthiness of the new batteries in 2007.
A year later, Boeing participated in an RTCA committee that established a set of more rigorous certification standards for lithium ion batteries on aircraft than the FAA had developed for the special conditions on the 787.
The NTSB wanted to know why the 787 was not required to meet the newer standards while the certification process was ongoing.
"The first thing to remember is that the RTCA standard is never a regulatory requirement," says Steve Boyd, the FAA manager of the airplane and flightcrew interface branch. He agreed that the RTCA standard was "more severe than in our special conditions. We think that's excellent that industry establishes a regulation for themselves that exceeds the regulatory standards."
Ultimately, however, the RTCA standards were focused on preventing failures at the level of the battery and the individual cell, he says. By contrast, the 787's certification standards were established at the aircraft level, so potential failures at the battery level could be offset by other systems at the aircraft level under the special conditions, Boyd says.
The RTCA issue was only one of several distinctions in the hearing drawn by Boeing and the FAA against the terminology used by the NTSB to describe aspects about the battery incidents.
For example, the witnesses clarified that there was never an "explosion" on the Japan Airlines or All Nippon Airways 787s that experienced battery failures, although NTSB officials used that term in public statements about the incidents three months ago.
NTSB chairman Deborah Hersman and other board members also queried Boeing and the FAA panelists over the exact level of safety demanded under the agency's special conditions for battery certification.
For example, the NTSB has said that the certification standards were supposed to prevent such a battery failure from occurring in less than one in 10 million flight hours, yet both incidents occurred before the 787 fleet reached 50,000 flight hours.
Sinnett said that the "one-in-10 million" standard applies only in very limited conditions.
"That number applies to a failure of a cell for particular reasons," he says, while listing installation damage as an example of a root cause that would be allowed to occur more frequently than under the special condition standard.
While admitted it was "serious" and "must never happen again, Sinnett also appeared to stand by previous comments made in Tokyo that neither of the 787 battery incidents in January involved the kind of catastrophic failure that the special conditions were designed to prevent.
Sinnett's comments were seconded by Ali Bahrami, the FAA's manager for the air transport directorate, which established the special conditions for lithium ion batteries and approved the 787's airworthiness certificate.
In neither incident did the battery failure present a risk to the overall aircraft, Bahrami says.
"The overall system worked," Sinnett adds. "We didn't have a catastrophic outcome."
Even so, the battery incidents and subsequent grounding have changed how Boeing will approach applying new technology in the future.
"In retrospect, we may apply tighter test criteria," he says, "or seek to understand the test criteria more in areas of new technology."