DID Boeing gamble more than it bargained for when it embarked on its 787 Dreamliner programme? The company bet heavily on not just one but three technological leaps into the unknown—any of which would, by itself, have been enough of a challenge for other aircraft manufacturers.

First, it opted to build more of the aircraft's structure out of carbon-fibre composite material (instead of aluminium alloy) than had ever been attempted before. Then, it took outsourcing to the extreme, assembling the plane from plug-in parts supplied by an unprecedented assortment of foreign and domestic manufacturers. Riskiest of all, it ditched the conventional hydraulic systems used for actuating a plane’s moving parts, and replaced them with electrical controls.

Going all-electric meant resorting to more powerful battery packs to provide standby power for when the plane’s generators were idled. Boeing decided that the only rechargeable batteries capable of doing the job, without incurring too much of a weight penalty, were lithium-ion cells. These have twice the energy density of the nickel-metal-hydride batteries used in hybrid cars like the Toyota Prius—and up to six times that of the lead-acid batteries used in conventional cars.

Overall, this wholesale departure from the tried and true was for one simple purpose: to save weight, and thus fuel—the largest of an airline’s direct operating costs. As a design strategy, it worked better than many expected. The 787 Dreamliner weighs considerably less than any comparable twin-aisle jet capable of carrying around 250 passengers and, as a result, uses 20% less fuel. As a bonus, the plane is reckoned to be 30% cheaper to maintain.

For airlines eager to reduce operating costs, the 787 Dreamliner thus promised to be every bit of a dream come true. Even while it was still on the drawing board, carriers and aircraft-leasing firms jostled to get places in the queue for early deliveries. To date, Boeing has booked orders for 848 Dreamliners, and has delivered 50.

This is not the first time Boeing's engineers have bet the firm on a radical rethinking of aircraft design. Back in the late 1960s, the company almost went broke developing the world’s first wide-body commercial jet, the Boeing 747. It prevailed and went on to sell more than 1,450 of them. It is no exaggeration to say that, with its huge passenger capacity, the jumbo jet fundamentally changed the economics of air travel.

Boeing hoped to do something similar with the 787 Dreamliner. But first it must prove either that the plane’s lithium-ion battery systems—two of which have burst into flames over the past few weeks—can be tamed, or, if they cannot, that they can be replaced with something safer.

Relying more heavily on electrical power than any other commercial jet, the 787 Dreamliner uses two 32-volt battery packs, containing eight lithium-ion cells apiece. These are not employed during normal flight, but are kept fully charged by the plane’s main generators ready to step in when needed.

Apart from being lighter than other rechargeable cells and able to operate at a higher voltage, lithium-ion batteries have no “memory effect” (the tendency to accept less and less charge each time they are recharged). They can also be charged faster than most other cells, and they hold their charge far longer.

The downside is that, if overcharged, physically damaged or allowed to get too hot, lithium-ion cells may experience thermal “runaway”—generating heat faster than it can be dissipated. A cell may then rupture, releasing inflammable gases that ignite and cause a fierce fire or an explosion.

Similarly, if moisture or other contaminants get into a cell during its manufacture, it can short-circuit and again trigger a fire or explosion. Draining a lithium-ion battery completely can also cause it to short-circuit, making recharging dangerous. For these reasons and more, all lithium-ion batteries contain sensors and circuitry that shuts them down when their voltage rises above or falls below certain levels.

Even so, lithium-ion batteries have caused numerous fires. General Motors, for example, had to offer to buy back all the Chevrolet Volt plug-in hybrid cars it had sold since the model was launched after several lithium-ion battery packs burst spontaneously into flames following tests.  Sjekk tillegg til denne artikkelen nederst.

Nor is it just big vehicle batteries that cause problems. In 2010, a UPS cargo plane caught fire and crashed shortly after it took off from Dubai. Fingers were pointed at the cargo of lithium-ion batteries intended for portable gizmos as the most likely cause. Earlier still, Sony and other consumer-electronics firms had to recall 10m lithium-ion batteries following a series of laptop fires.

The Federal Aviation Administration (FAA) has a list of more than 100 cases of fires on board aircraft that were sparked by lithium batteries. As for using them in commerical aircraft as part of the equipment, that was long considered too hazardous. Lately, however, the FAA has been giving plane-makers consent to do so on a case-by-case basis.

For the 787 Dreamliner, Boeing was required to install four layers of protection to prevent a short-circuit in any one of the battery’s eight cells from affecting others. A pressurised air system was incorporated to carry smoke and toxic gases away from the cabin. The company felt confident that, if a cell were to experience thermal runaway and catch fire, the flames would be contained safely within the battery-pack’s enclosure.

That does not seem to have been the case. In the two fires aboard 787 Dreamliners—first, on January 7th, after a Japan Airlines flight landed at Boston and the passengers and crew had disembarked; second, on January 16th, when an All Nippon Airways flight was forced to make an emergency landing in Japan after smoke filled the cockpit—the multiple protection systems on the aircraft failed to do their job adequately. All 50 Dreamliners currently in service were subsequently grounded indefinitely, and deliveries of new ones halted.

So far, air-safety investigators in both Japan and America agree that, in neither case, was there evidence of the batteries being overcharged. The flight recorders show their voltage was correct before the fires broke out. That would seem to rule out the charging system as the source of the problem. By the same token, it would suggest the battery-management system, which is used to keep the voltage within its prescribed limits, was working properly.

The investigators also agree that the fires cannot be put down simply to a faulty batch of batteries. Their serial numbers suggest these came from different lots. It is therefore unlikely that manufacturing defects caused the short-circuits that made them overheat and catch fire. Unfortunately, with the fires having been so intense, any evidence of a fault lying in the actual wiring of the battery-management systems went up in smoke.

Faulty wiring is a common cause of overheating in electronic devices. If the connections are not rugged enough, components can get sufficiently hot to make sparks fly. Should a lithium battery be in close proximity, the heat could trigger thermal runaway in one or more of the cells.

General Motors faced a similar quandary with the Volt's faulty battery pack. The conclusion was that malfunctioning sensors, rather than chemical reactions going haywire within the cells themselves, were the source of the problem. The solution was to laminate the management system’s circuitry, beef up the battery pack’s cooling lines, and reinforce the tray containing the battery modules.

That is probably the least Boeing will have to do to satisfy the FAA. But more extreme measures may be needed. Ultimately, that might mean abandoning lithium-ion batteries altogether and replacing them with nickel-cadmium (ni-cad) ones. That is what Cessna was forced to do in 2011 after the lithium battery in one of its Citation CJ4 business jets caught fire. Such a move by Boeing would keep the Dreamliner grounded for possibly as long as a year, as the plane’s electrical system was redesigned and resubmitted for certification.

Boeing could have avoided its current woes had it adopted ni-cads in the first place—or, at least, heeded recommendations for more stringent testing of lithium batteries made in 2008 by RTCA, an independent standards body that advises the FAA. Both Boeing and the FAA chose to ignore the tougher recommendations for fear of delaying the 787 Dreamliner still further. Instead, to save weight, Boeing gambled on the powerful lithium battery, knowing full well its risks. The irony is that, in doing so, all it saved was 18kg (40lb) per plane—about the same, one expert noted, as a single piece of baggage.

Tilllegg:
Here's the problem: While the battery cells in Boeing 787s and, say, Chevrolet Volts are both in the lithium-ion family, they use very different chemistries. You can think of lithium-ion cells rather like motor vehicles: They all do some variation of the same thing, but there are many different types, sizes, shapes, and different technologies to make that happen. Consider the difference between gasoline and diesel engines, for example. The cells in the 787, from Japanese company GS Yuasa, use a cobalt oxide (CoO2) chemistry, just as mobile-phone and laptop batteries do. That chemistry has the highest energy content, but it is also the most susceptible to overheating that can produce "thermal events" (which is to say, fires). Only one electric car has been built in volume using CoO2 cells, and that's the Tesla Roadster. Only 2,500 of those cars will ever exist. The Chevrolet Volt range-extended electric car, on the other hand, uses LG Chem prismatic cells with manganese spinel (LiMn2O4) cathodes. While chemistries based on manganese, nickel, and other metals carry less energy per volume, they are widely viewed as less susceptible to overheating and fires. So if you see coverage of the Boeing 787 battery fires that says anything at all about electric cars, do consider dropping a friendly note to the reporter involved. It may be unreasonable to expect every reporter in the world to know that "lithium-ion batteries" are a family of very different chemistries. Science reporters, on the other hand--let alone engineering professors--really should know better. You have been warned.