In Search of the Flux Capacitor - Part 2
I GOT THE POWER
Jay Leno's voluminous garage suggests why. "I have a 1909 Baker Electric," boasts the coiffed comedian, who's as renowned for his car collection as his Woganesque humour. "That goes 110 miles on a single charge." What's more, it still works. "I have never done any maintenance," he insists, "other than maybe greasing the wheel hubs." To consumers reared on built-in obsolescence, to say nothing of the business plans depending on it, this can come as a shock. "You don't do anything," Leno repeats. "You plug it in, charge it and drive it. The motor is virtually maintenance-free." Apart from the Tesla, which makes similar boast about servicing, there are few cars in production or in the pipeline that can match the Baker's mileage, even if they could run it off the road. The Th!nk City, for example, due like half a dozen other mini EVs to hit U.S. tarmac by 2010, claims to get 124 miles per charge, though it would be less if you drove flat out at 60 mph.
Not much had changed in battery technology when the EV1 was launched. Its lead acid cells used 100-year-old science and although fairly effective they were heavy and far from ideal. The nickel-metal hydride (Ni-MH) replacements weren't particularly light either, but performance improved quite a bit and some Japanese cars still use something similar. Switching to lithium cells helped shrink mobile phones and the same principle delivers cars more power from less weight, although the Tesla Roadster's half-tonne bulk in the boot does have the advantage of generating more downforce, which comes in handy when navigating tight corners. The obstacle for many years was lithium's reactivity, which is why it can store so much energy. It's enough to "incinerate anybody in the car," warns Barrie Lawson of Axeon, a Dundee-based battery maker that's joined up with Allied Vehicles to launch electric taxis in London. "That's always been the great fear."
In the case of the Tesla, which has passed U.S. crash tests, there are 6,831 little cells to worry about, lined up in 11 separate sheets and grounded in a gloopy substance to keep them apart. Even if one blows it won't affect the others, the company says, so there's nothing to worry about when driving. Even if you crash, and a battery's broken up, an open circuit cuts its current instantly, says Barrie Lawson. Still, a spate of spontaneous laptop combustion spread panic among carmakers. Manufacturing safety checks have been tightened, and Toyota postponed plans to introduce new battery chemistry for the Prius. Lawson cautions against reading too much into this, stressing that the danger from electric parts dislodged in an accident would be "no worse than spilling a tank of petrol." Moreover, he says, "if someone came along today and proposed that you carry this highly inflammable stuff around just with a tank with a thin steel wall around it, they would say you were crazy."
Martin Eberhard concurs. "To me it's actually quite amazing that you can take a relatively small car, smack it into a truck at you know, 50 or 75 miles an hour and the gasoline tank doesn't just explode all over the place every time." Thankfully the same seems to be true for the Tesla, although rival carmakers still make scoffing noises about the number of volatile batteries joined together. In one sense they sort of have a point: the greater the number of cells, the harder it is to ensure that they hold a full charge without them automatically defaulting to the storage capacity of the weakest in the series, which weakens further every time you load it to below maximum. "We've got 72 cells in our battery pack," says Greg Starns, the head of software development at Frazer-Nash Research, another firm which plans to launch electric taxis in London this year. "And we're doing everything we can to make that fewer."
The problem is called equalisation and the solution a battery management system, or interface, which every EV needs to run effectively. The Tesla's was "a huge challenge", says Owen, though he claims it's no longer a concern – the car will load up from its wall-mounted three-phase charger in just three or four hours. Sceptics crunching these numbers say that it might take more than five times longer, at best, using a regular plug socket, but Tesla says an overnight charge is still doable. Nevertheless, this battery set up alone costs between $20-30,000, which is what American consumers are used to paying for a mid-range SUV. Put it in a heavier car, like Tesla's new Model S sedan (which may turn out to be a four-door hatchback if rumours are true), and performance would suffer. Even that vehicle, due in 2010, is expected to sell for somewhere above $60,000. A mass-market model at half that price is still just the stuff of Tesla fantasy, although it's talked about as next on the to do list.
"It all comes back to the battery," says Bill Reinert, of Toyota's U.S. Advanced Technology Group. "If you want to run longer and further on electric power alone, it means a bigger battery, it means charging a battery more fully and discharging it more completely. And it means provisions for cooling or ventilation in order to give the batteries longer life." That means choosing what you want and what you're willing to give up, in terms of space, weight and convenience as well as price. Glyn Owen expects batteries to go like microchips, though his EV version of Moore's law is a more conservative 10-year forecast of doubled power and halved size. "There's room for at least another doubling of capacity," agrees Martin Eberhard, who still holds a chunk of Tesla stock, if not management responsibility for talking up the company's prospects. Barrie Lawson isn't convinced. "There's nothing going to come along and make the battery half the size in the near future," he says. "There's no huge breakthroughs on the horizon that will make a big impact on the chemistry but there are many variations on the chemistry to optimise the battery performance."
It's a question of "flavours of lithium", to use Lawson's favoured phrase. Crudely simplified, batteries use lithium in its ionic form (which means stripping the atoms of an electron to make them positive). When the cell's fully charged, these ions congregate around its anode, which is usually made of graphite. During use, ions migrate within the battery to the other electrode, the cathode, and negatively charged electrons pass between the two via an external circuit that drives the motor. Electrodes are where it gets interesting.
Early cathodes were made of expensive, and scarce, cobalt oxide. Manganese oxide is getting popular now, but more attention has been focused on iron phosphate, which has less of a tendency to overheat. Making lithium cobalt (shorthand for nickel cobalt aluminium cathodes) safer cut its range by 30 percent, Lawson says, so it's all a question of balancing different priorities. Switching from carbon anodes to lithium titanate allows for faster charging at lower temperatures but available voltage drops by a third or more. Lithium iron phosphate cathodes don't offer the same speedy charging benefits but they do allow for usage across a wider range of charge; between 10 and 100 percent of capacity, as opposed to between 30 and 70 in most batteries, to minimise the danger of side-effects. Lithium iron phosphate comes cheap and although its performance isn't as good at lower temperatures, and isn't always so easy to monitor, it's the option of choice for many plug-in hybrids, including the Volt.
After five years of mugging up on batteries for Tesla, Martin Eberhard isn't convinced. "I'm actually fairly pessimistic about lithium iron phosphate," he says, although he expects it to be a short-lived hit. "They're already approaching the theoretical limit of the electro-chemistry." Lithium manganese has more potential, he argues, when you start tinkering with its crystal structure, and the surface area on which ions gather, increasing potential capacity. The key to this is nanotechnology, which can also help speed up charging times. Altairnano, whose batteries are used in the Lightning, claims it's working towards a 10-minute charge time, from 480-volt outlets that could be placed in a roadside service station. Other researchers say it can be done even quicker, but there's a risk of overheating. Gerbrand Ceder at the Massachusetts Institute of Technology runs a project to find new combinations of materials that can address this problem. His team has modified a lithium iron phosphate battery to make it a quarter the average weight, and way faster to charge. "We can take all the power in or out of our battery in 10 seconds," Ceder says. "You put that in a Prius and it accelerates like a Ferrari."
That's all still stuck in the lab, though perhaps not for long. What's already in productive use is extended longevity. A123 Systems pioneered this with a battery that, as Eberhard puts it, "although lower charged, went much, much more cycles, and the rate of loss of capacity with each charge cycle was fairly low. But today, if you look at Sony's iron phosphate, and some of the others like Sanyo, they're already a better battery than the A123." This is the source of his projection for manganese. "If you could take a lithium manganese cell," he says, "and without increasing its capacity at all, just simply increase its cycle life by a factor of two, you'd have made a phenomenally good battery." One way or another, he thinks it's inevitable. "Capacity of batteries for the same volume and the same mass has increased historically 8 percent per year; that's been true for 25 years now," he says. Cost per cell has dropped in the same time frame by around five percent a year, though it's stabilised since 2006 because of surging demand. "I'm optimistic that, over time, the batteries will get cheaper, and will get higher capacity in the same volume," Eberhard concludes. "The venture community here in this country has been investing like crazy in battery technologies."
Then there are design issues to consider: flat or prismatic being better for saving space (which matters more in a car than in a truck), while the winding cylindrical variant can in theory store more energy, but allows for less space between cells, decreasing its advantage. In any case prismatic batteries come in multiple forms: stacked as plates (which maximises energy), wound around like a Swiss roll (which cuts cost) or packed in a pouch (which is cheaper still but more vulnerable to damage, so has to be more densely packed). It's all about priorities again: "The best battery for the Tesla is not the same as the best battery for the Volt," Eberhard says. "When people say what's the best battery, I would ask ‘for what'?"
Whatever you want to do with batteries, you'd be hard pressed to do it outside China. There are at least 100 manufacturers of lithium cells there, Lawson says, and some say more like 4,000, though most of these are just wholesale traders. "Only about half a dozen or a dozen are any good," he says. "And there's no loyalty in China. Everyone in that industry knows each other and they're switching firms all the time and taking the technology with them." All of which may help drive improvements, but you need to know where you're buying, he cautions, having written a guide to the pitfalls. "It's no use expecting support from any of these companies unless you turn up at their door with a multi million dollar purchase order," this document stresses. "You will be competing for their manufacturing capacity with the world's major battery consumers. Orders under $100,000 are a nuisance."
Competition and capacity are tight enough to squeeze the industry's growth prospects, just as DRAM bottlenecks once held up PC development. While some question how much lithium there is in the world, battery companies say it's just a question of recovering more of it. "We know it's very abundant and it's even in the sea," says Charles Gassenheimer, the CEO of EnerDel. "Maybe we'll have to figure out how to extract the lithium from seawater." Other raw materials are required in greater quantities anyway; lithium's a relatively minor ingredient. But that doesn't mean there won't be problems. "If this market goes to a trillion dollars a year there will be bottlenecks," Gassenheimer scoffs. Even the most ambitious forecasts are around a 20th of that, but a more realistic prospect does worry him. Like all the major battery makers he's talking to all the big car companies and they all have the same question: "Can you get us scale?" To which he says: "I can't get you scale until I build capacity," which he can't do "until I have a volume order," which won't come without proof of scale. The solution? Either "the federal government gets involved", or "the car companies kick in some money". Otherwise there's a shortage in the offing, though Gassenheimer sees another reason not to worry: he thinks conversion of the world's cars to electric power will take at least 30 years.
PLUG-IN OR COP OUT?
For all the attention lavished on Toyota's Prius, it's not the world's most fuel-efficient car. "I get better mileage in my M5 BMW diesel," says Greg Starns at Frazer-Nash. Until now, the better-performing diesels have seemed the best bet for drivers anxious to cut costs, yet concerned about limited EV range. Rising prices for liquid fuels have changed the equation, if not yet driving habits, which means people still prize freedom over weaning themselves off petroleum. "The U.S. is 3,000 miles wide with huge tracts of sparsely populated land," notes Sherry Boschert, a journalist and pro-EV activist, who's notched up 66,000 gasoline-free miles in the past six years. What's more, "our mass transit system sucks." Since most car-owners live in suburbs, their vehicle is essential to daily life. Yet statistics show they don't drive as far as they sometimes think. Existing EV range is adequate for around four out of five trips. A Department of Transport survey in the 1990s found that half of all motorists in the U.S. travelled 25 miles a day or less, and 80 percent drove fewer than 50, well within the single-charge range of the earliest General Motors EV1. Last year, the average U.S. trip was 9.9 miles and the average daily drive was 32.7. In Britain, the figures were more like 8 miles a trip and 10 a day. The median (which isn't published) would probably be even lower still.
"The mass market won't always be wary of full EVs," says Boschert, who's written a book to entice people to buy one and formed a campaign group to lobby companies and lawmakers. "That wariness is more a figment of the automakers' imagination than reality. Tons of people would buy an EV right now if one was available. We hear from them all the time." Some of them have already, and they report fascinated interest from passers by. Todd Poelstra's all-electric Zap Xebra has a number of obvious drawbacks: it only does a couple of dozen miles and the top speed is limited to 25 mph in many states, because it's technically classed as a motorbike. Oh, and it's only got three wheels. And it comes in "look at me" lime green, which mightn't be quite so bad if it didn't look so much like a 21st century Reliant Robin. Yet when someone passes him in the car park shouting "Right on!", Poelstra tells an American radio reporter that this sort of thing is a daily occurrence. "We've had our picture taken countless times," he says. "Every time we park, somebody wants to talk about the car."
Fewer than 1,000 Zap Xebras have been sold in the U.S., which is roughly the number of EVs trundling around London, most of which are Indian-made Reva G-Wizes, though there's increasing competition from the more attractive Mega (made by a company called No Internal Combustion Engine: NICE). From roughly $12,000 for the Xebra to £12,000 for the Mega, independence from fossil fuels is both affordable and available in multiple options (assuming you don't burn coal or natural gas to power the grid that charges them). But you'd have to pay as much again for a battery that goes 100 miles, which is partly why people are holding off.
Yet offer them the chance to do most of their trips on electric power, without removing the option of filling up for longer journeys, and everything changes drastically. "This is not a pipe dream," says MIT's Gerbrand Ceder. "The battery technology is there to make this happen. GM, Volvo, Mercedes, Nissan – they're all working on it." Companies are falling all over each other to talk up plug-ins and with fuel economy requirements getting tougher, it's little surprise that one survey (by Roland Berger and J.D. Power) forecasts half of Europe's cars could be hybrids by as early as 2015 (though what proportion would be plug-ins isn't clear). This outcome is no less likely across the Atlantic, even if later. "The plug-in concept is so important because driving patterns in the U.S. support it," Ceder says. "With even a 20-mile range, you can do 50 percent of all your driving."
From an efficiency point of view it's a no-brainer. While there's no liquid fuel that's as densely packed as oil or LPG, the internal combustion engine's bad at exploiting it. According to the U.S. Department of Energy, less than 20 percent of petrol propels a vehicle; the rest just gets wasted as heat. In electric motors, by contrast, 86 percent of the energy converts to power. Traditional hybrids like the Prius save fuel by shutting down the engine when stationary or moving slowly, recovering energy for future use through regenerative braking (which captures kinetic energy from the engine, turning it into a generator as it slows), and reducing the engine's size to run it more efficiently by combining it with computer-controlled battery power.
Plug-in hybrids are different: they run on batteries all the time if you stay in range, which most drivers already would on their daily commute (assuming the cars live up to promises). Effectively, they're EVs with insurance, which is why the verdict of Paul Nieuwenhuis, an automotive expert at Cardiff business school, is almost universal: "The plug-in hybrid will be the winner," he says, "although I also forecast an increase in battery EV demand." Others expect the former to fuel the latter. "Give the worldwide consumer five years of driving a plug-in hybrid," says Martin Eberhard, "and some of those drivers will realise that they actually never use the range extender, and the next car they buy won't need it."
Not everyone's so sure, including some major manufacturers. No one's actually built a plug-in hybrid yet; the concept only exists because of home-made hacks for existing cars like Priuses. The fundamental drawback is fitting two substandard drive-trains to every vehicle: an engine that no longer works at peak efficiency (the aim of the Prius-style hybrid) and a battery that can't take you far enough to ditch the engine. That means lugging it around without using it most of the time, along with all the additional bits of kit to run it smoothly. And when you're forced to switch to petrol you've got a useless electric powertrain weighing you down and an under-capacity engine having to pull it.
Although campaigners have slated Ford for pouring scorn on plug-in hybrids, it's probably just being franker than most of its rivals. "We are working on the technology," a spokesman says, "but we have to determine: how well does this hold up in the real world?" Defending the decision to hold off for the time being (a launch "between 2012 and 2020" is the closest the company gets to making a commitment), Ford's manager in charge of energy storage is even blunter: "If customers aren't buying them, we're not making them," Ted Miller says. "If there's going to be a true plug-in hybrid market, we're going to be there. It's just that that's a huge commitment to actually go to production."
This is the sort of thing weighing on Sherry Boschert when she asks if carmakers will really deliver. "I'd like to believe they'll give us plug-in vehicles, but until I can go out and buy one, I'm still sceptical." Another campaigner is more forthright. "The corporations are sitting, wishing this whole friggin' thing to go away," says this publicity-hungry activist. But who is he? Ralph Nader? Michael Moore? No. It's the former Chairman of Intel, Andy Grove, who says this reflects "exactly what the computer companies' attitude was to personal computers." Based on this analogy, by which hobbyists have to show bosses there's really a market, Grove is lobbying the government to help retrofit America's least fuel-efficient vehicles, starting with 80 million gas-guzzling pickups and SUVs. Unsurprisingly, the companies remain to be convinced, not least GM, which sells many thousands of these vehicles every year and spends millions on dreaming up new ones, though they never seem to get much more efficient. "We strongly discourage consumers from retrofitting vehicles," a GM spokesman says. Fancy that.
Nevertheless, it's largely making the Volt because of what conversions have proved can be done. It's not quite as simple as buying a generator and bolting it onto your Prius, but it's not all that far off: the first plug-in "Prius+" was developed by two Californian activists, who, in the words of one, Felix Kramer, simply "figured out how to hack it." There was an unmarked button on the car that allowed drivers to switch to all-electric mode for the car's one-mile capacity, but it was disabled on U.S. models because of complicated emissions-testing rules. Crowdsourcing help from dozens of engineers over the Internet, Kramer and his friend Ron Gremban retrofitted a car to go 10 miles on nothing but battery power. "It's not that big a leap in technology," one of their backers said at the time. "This isn't rocket science." The trouble is that you invalidate the warranty, and the liability for meeting emissions laws. Undeterred, companies in the U.S. now offer retrofits for all kinds of models, but a pure electric conversion can cost over $30,000. Even a plug-in hybrid kit, available from A123, will set you back $10,000. In the UK, a company called Liberty offers revamped Range Rovers for £95,000, with promised range of 200 miles and performance on a par with standard versions.
It all gets positive media coverage, but plenty in the industry aren't so sure. "You can't make a real company out of converting somebody's cars," Martin Eberhard says, mainly because of the liability implications. "In this country, an automotive manufacturer is required by law to warrant the entire emission system of the car for 100,000 miles, and it's probably the same where you are. And once you change the electronics, the computer that's controlling all that, Toyota is off the hook, and suddenly you are on the hook; you are liable. So, you can do that for your car and mine and your friends', but if you make a business out of it, you've got a problem." Whether this is really true remains to be seen. If not, it could cannibalise demand.
TO BE CONTINUED...
blog comments powered by Disqus