My copy of the book is in storage or I'd give numbers

Post #235,124 by ben_tilly 11/19/05 12:52:24 PM Reply	My copy of the book is in storage or I'd give numbers But yes, you are right that this assumes the future existence of far better batteries than we have today. However that is more reasonable than you might think. It turns out that Moore's law is not an isolated example. Most technologies that have an obvious metric to be measured by show exponential improvement on that metric over fairly long periods of time. The exponents tend to be smaller than in Moore's law, but they are there. The 2020 estimate was achieved by looking at the performance of batteries in 1997, and looking at the apparent rate of improvement in energy density and discharge capacity. Given the minimum requirements for a US car (sufficient acceleration to get onto an Interstate, good enough braking for emergencies, able to drive at least 100 miles before refueling), battery technology won't be good enough until around 2020. So sure. Batteries are not good enough today. Have not been good enough for the last 30 years. And won't be for many years to come. But unless batteries stop improving, they will eventually become good enough. In fact a clear sign of how much batteries have improved is the viability of hybrid cars. We could have built hybrid cars a decade ago. But there was no point - batteries weren't good enough for you to get enough out of the complexity to be worthwhile. Today they are good enough. And to answer the common objection of whether electric is better than gasoline, just look at the mileage that hybrids get. Electric has a huge advantage over gasoline - you get to recover a lot of energy in normal driving conditions. That's why hybrids get better mileage than gasoline. And pure electric has several advantages over hybrids. They have a simpler design, lighter engine, are more efficient and more reliable. (All of these things come from not having 2 engines with complex subsystems to switch between them as needed.) There is no question that, if you could build a good enough electric car, it would be better than what we have. The technical problem is that today you can't build that car because batteries aren't good enough. The business problem is that when electric cars are barely good enough, traditional car manufacturers won't know how to market and sell them. And by the time they are really good enough, traditional car manufacturers will be facing entrenched competitors whose pricing structure is much leaner. Cheers, Ben I have come to believe that idealism without discipline is a quick road to disaster, while discipline without idealism is pointless. -- Aaron Ward (my brother)
Post #235,141 by Another Scott 11/19/05 2:24:12 PM Reply	This paper from MIT in 2000 is interesting. [link\|http://web.mit.edu/6.933/www/Fall2000/solarEclipse.pdf\|Solar Eclipse: The Failure of a Promising Technology]. It's about the history of a solar powered car that was being developed at MIT and talks about it in relation to Christensen's book. p.23 Stepping back from the cars themselves and studying their history, trajectory, and demise have brought to light the indicators to their failure, the survival skills of those involved in designing them, and their impact on technology today. The initial belief that these cars were going to cause a real threat to the gasoline industry slowly yet surely disappeared. The engineers could only push the technology so far before the physical limitations of the cells and the sun's energy were too large to overcome. By the time the parts had become reliable, the whole was still not sufficient. There was no way to put the components together such that a customer would switch from gasoline to solar. This study thus shows that such technologies fail because of exactly this situation: the innovations saturate before a viable product can be produced for the targeted market. The question is then put forth of whether there could have been a better market, and that is a possibility that remains, yet cannot be determined until someone attempts it. The MIT team went on to found [link\|http://www.solectria.com\|Solectria] which is now (or part of) [link\|http://www.azuredynamics.com/\|Azure Dynamics]. They still make all electric vehicles (apparently no longer solar powered), but they also do a lot of hybrid vehicle work. Ben writes: So sure. Batteries are not good enough today. Have not been good enough for the last 30 years. And won't be for many years to come. But unless batteries stop improving, they will eventually become good enough. It depends on what you mean by "good enough". And that's obviously where the choice of market comes into play. The military is willing to spend millions on electric vehicles that have unique capabilities. Box, e.g, probably isn't. :-) Will electric-only vehicles displace hybrids? In some niches, yes - it already does (see below), but not overall. As Ashton pointed out, the power density in liquid fuels is much higher than that obtainable from batteries. There will continue to be a need for internal combustion engines for a long time to come (e.g. you're not going to see a practical electric 18-wheeler anytime soon.) Meaning several decades, IMO. Car companies are continuing to improve the efficiency of [link\|http://en.wikipedia.org/wiki/Otto_cycle\|Otto-cycle] internal combustion engines. VW has a new turbo+supercharger [link\|http://www.autoweek.com/news.cms?newsId=103408\|"Twincharger"] engine that gives good performance and improved fuel economy. Improved electronic controls allows reduced emissions and improved economy in even conventional cars. Even some [link\|http://www.hybridcars.com/silverado-sierra.html\|large pickup trucks] now have features like turning off the engine at a stoplight to improve mileage and reduce emissions. Conventional cars are going to become more hybrid-like because it's better (from emissions and efficiency standpoints at the moment; potentially from a total cost standpoint) to convert motion to stored energy than generate fresh motion. Even electric cars have or will have regenerative braking in most cases. Yes, batteries are going to improve, but they've been worked on for a very long time (Franklin coined the term in 1748) and the improvements are coming at greater and greater cost. Several other electric energy storage systems are [link\|http://www.onr.navy.mil/fncs/aces/focus_pwrgen_energy.asp\|under investigation] but no breakthroughs are in sight yet. So, at least for the next 50 years (my wild guess), it's probably safe to say that conventional autos will be around and a large part of the market, but they'll become more hybrid-like over time. I also think it's safe to say that companies like Toyota, Honda, Hyundai, BMW, Daimler-Chrysler and VW (or their successors) will be providing most of the vehicles - not Azure Dynamics. IOW, I don't think electric vehicles are a good example of a disruptive technology - as I understand the term (and with the caveat that I've only skimmed TID). FWIW. Cheers, Scott.
Post #235,144 by ben_tilly 11/19/05 3:02:53 PM Reply	What I mean by "good enough"... is good enough to make a low-end all-purpose consumer car. At that point it will be possible to deliver a reliable all-electric car at a much better price point than an internal combustion one. It will be a long time - maybe never - before they are good enough to match internal combustion engines for high-end purposes. But they don't have to to win in the market. They just have to meet people's actual needs at a better price. And the complexity and overhead of doing internal combustion and other stuff means that internal combustion won't be able to match the price of electric. But you're right. Until it happens, this is not a good example to use. If it pans out, it will be a very good prediction. But for examples it is better to use something that already happened. An incidental point. When disruptive innovations start disrupting, the high-end market is always stuck on the old technology. You see, even as the disruptive innovation becomes good enough for many, it still sucks relative to the established technology and is not good enough for high-end use. Take, for instance, the case of hydraulic shovels. Starting in the 50's, the traditional wire shovels started being disrupted by hydraulics. But as of 1997, there were still 4 wire shovel companies left. However they had retreated to the top end of the market. In fact the only market that was left was building scoops for strip mining. If you need to remove less dirt than an entire hillside, hydraulics are good enough. But they can't do that (yet). Take, for another instance, traditional steel mills vs mini-mills. As Christensen spends some time explaining, the steel industry is segmented according to how demanding the metal is to produce. The mini-mills have been working their way up to harder and harder metals. Yet despite the fact that they have been eating away at Bethlehem Steel since the 70's, the highest-end steel production still is done by Bethlehem. Cheers, Ben I have come to believe that idealism without discipline is a quick road to disaster, while discipline without idealism is pointless. -- Aaron Ward (my brother)

Welcome to IWETHEY!