In-flight fire compounds Boeing's 787 challenges

The latest setback to the Boeing 787 programme could see entry-into-service pushed back as far as 2012, writes Ian Goold.

13th Dec 2010


The closing days of 2010 should have marked more than 30 months of Boeing 787 commercial operations by launch customer All Nippon Airways (ANA). Yet as Asian Aviation went to press the new twin-aisle twinjet remained a troubled project, uncertificated, with the test fleet grounded as analysts and customers expected the announcement of a seventh formal programme delay.

Boeing was unable to say in late November when flight-testing would resume, after an in-flight electrical fire on the second 787 test aircraft (ZA002) led to an emergency landing in Texas earlier in the month. The US manufacturer still awaited the results of initial investigations into the event, with at least one Asia-Pacific customer – Australia's Qantas – reportedly understanding that there had been a “significant" problem and that deliveries would slip to the end of next year (see box).

Likewise, Air Lease chief executive Steven Udvar-Hazy – who had been the 787's largest customer in his former role as chairman of International Lease Finance Corporation (ILFC) – was saying first deliveries would “definitely” be postponed: “It’s a big setback for Boeing," he said.

Udvar-Hazy’s remarks came a day after Morgan Stanley aerospace analyst Heidi Wood had said the first delivery "could" slip as far back as 2012, with the second half of next year the most likely timeframe for ANA to receive the new aircraft. Wood said she expected test flying would not resume until at least well into December. Wedbush Securities analyst Kenneth Herbert suggested a delivery delay until June or July.

Share impact

A prospective delay had already been accommodated in the pricing of Boeing stock before the fire, so there was relatively little volatility. Nevertheless, Boeing shares offered on the New York stock exchange were seen as providing "a buying opportunity because so many investors already are braced for a delay," said Alex Hamilton, managing director of Early Bird Capital. "There's no certainty as to when flight-testing is going to resume. Wall Street hates uncertainty."
By definition, Boeing's best forecast for when deliveries might begin applies only to the first aircraft. Given the circumstances that drove August's re-scheduling to early 2011, it was already likely before November's fire that outstanding work on early production aircraft would mean subsequent deliveries would be held up.

Boeing told analysts that the 20-30 aircraft parked around Seattle's Paine Field were "in various stages of final assembly" and that deliveries would "take longer than expected, particularly those with the Rolls-Royce engine". RBC Capital Markets aerospace analyst Robert Stallard suggested that following the handover of the first ANA 787, the delivery ramp-up is likely to be "longer and shallower" than previously anticipated. He said he believes only about 25 aircraft will reach customers next year, compared with earlier projections of as many as 80 deliveries in 2011.

Technical issues awaiting solution included identification of the cause of the earlier Rolls-Royce Trent 1000 engine failure during test in the UK, correction of errors in the build-up of the Alenia-supplied 787 tailplane (required to prevent premature fatigue), and reported problems with instruments.

In late November, Boeing was understood to be working with the US Federal Aviation Administration (FAA) to consider how at least five of the grounded test aircraft could be returned to flight while investigations continued. Two 787s involved in remote testing were later flown back to Seattle. Prototype 787 ZA001 was being refuelled in South Dakota when the fire occurred aboard ZA002 Boeing decided to forgo additional flights. The fifth aircraft, ZA005, was in California.
The incident flight involved two FAA people among the 42-strong crew: a pilot, who was flying the 787 from the left-hand seat at the time of the fire, and a systems engineer on board to observe during the more than six-hour flight. Failure of the electrical-power panel was noticed as the 787 passed through 1,000ft above ground level while preparing to land at Laredo.

Warning messages

Multiple warning messages reportedly appeared on flight-deck displays. After load shedding, the information is understood to have been confined to a single display on the left side of the instrument panel. This is likely to have determined whether the aircraft was landed by the FAA pilot or a Boeing test pilot (acting as the flight commander) in the right-hand seat.

Within days of the incident, investigators had developed a detailed understanding of the fire, although Boeing said more work was required to complete the investigation. Initially the manufacturer was able to say that the fire had caused ZA002 to lose primary electrical power. "Backup systems, including the deployment of the ram-air turbine, functioned as expected and allowed the crew to complete a safe landing," the company said.

The flight crew had retained positive control of the airplane and were said to have had "all of the information necessary to perform that safe landing". Initial inspection indicated that a power control panel in the aft electronics bay would have to be replaced. Boeing inspected the power panel and surrounding structural area to determine if other repairs would be necessary.

Flight data was retrieved from the aircraft and sent to Seattle for "several" days' analysis. The incident occurred as the flight-test team was monitoring the 787's nitrogen-generation system.

Boeing continued to conduct ground testing until the 787 fleet could be returned to the air and while it gauged the effect of the lost flight time. "We cannot determine the impact of this event on the overall programme schedule until we have worked our way through the data. Teams have been working through the night and will continue to work until analysis is complete and a path forward is determined."

In a later statement, Boeing reported that the "total duration" of the incident had been less than 90 seconds, with the fire itself lasting less than 30 seconds. After a complete inspection of ZA002, investigators began to prepare to install a new electrical-power panel and new insulation material. Unspecified "minor structural damage" was to be repaired using standard repair techniques in the 787 structural repair manual.

Boeing claimed the on-board fire demonstrated many aspects of the 787's safety and redundancy. The manufacturer said that, before any decision could be taken on resuming flight-testing, it had to complete the investigation and assess whether any design changes were necessary. "Until that time, Boeing cannot comment on the potential impact of this incident on the overall programme schedule," it said.

Leaked images


Within days, images of the fire-damaged interior of the 787's equipment bay began to appear on the Internet before being quickly removed at the request of Boeing, which indicated their proprietary nature. One on the Plane Talking website showed a charred thermal blanket and a melted electrical component, while another illustrated what was described as "destruction of alloy components" in the fire, despite its apparent brief duration.

Following removal of the images, Plane Talking posted a notice: "The publication or suppression of information about a fire on board ZA002 is not in itself going to affect the resolution of issues arising from the fire, nor the time taken to achieve certification. It is the fire that matters, not the management of the message."

As Boeing prepares to resume 787 flight-testing, Japanese programme partner Fuji Heavy Industries has begun to expand in anticipation of increased series-production rates. It is enlarging its factory in Haneda, where it broke ground on the expansion on 21 November. The factory, its third on the site, will handle integration of the 787’s centre-wing box and main landing-gear bay. Production is scheduled to start in November 2011.

Despite the recent hiatus in 787 completion, with suppliers having been asked to slow down (or even stop) output of components while Boeing catches up with final assembly or modifications to completed airframes, the manufacturer has taken on almost 1,000 workers in recent weeks. Alongside increased flow on the 787 line, Boeing is raising manufacturing rates of the 737, 747, and 777.


Boeing 787 timeline

The following key points are taken from a 787-programme timeline developed by Reuters news agency:

June 2007 – Boeing said its scheduled late-August first- flight date might slip to September, absorbing much of a one-month "window" that would still permit May 2008 first delivery.

27 July 2007 – Less than three weeks after 787 roll-out, Boeing acknowledged the programme was running slightly late in some areas, but held to the schedule.

September 2007 – Boeing delayed first flight about three months to mid-November/mid-December but the delivery schedule was unchanged.

October 2007 – Boeing confirmed an extended delay to late March 2008 and postponed first delivery by about six months.

January 2008 – Further three-month hold-up due to supply and assembly problems put back first flight to late June, with delivery forecast for early 2009.

April 2008 – Boeing re-scheduled first flight to the last quarter of year, and first delivery to 2009's third quarter.

November 2008 – Maiden flight delayed to 2009 by 58-day production workers' strike.

December 2008 – First flight rescheduled to April-June 2009, with delivery forecast as first quarter of 2010.

August 2009 – Boeing put back first flight to the end of year, with delivery in October-December 2010.

15 December 2009 – First flight of 787.

August 2010 – Boeing postponed initial delivery to mid-first quarter of next year.

September 2010 – ZA001 grounded for engine change following Trent 1000 failure at Rolls-Royce.

9 November 2010 – all six 787 flight-test aircraft grounded after fire aboard ZA002 leads to unscheduled landing.

16 November 2010 – 787s ZA001 and ZA005 cleared to return to Seattle from remote flight-testing elsewhere.
 

Asian Aviation at a glance