[extropy-chat] Car of the (near) future
hal at finney.org
Mon May 23 20:33:18 UTC 2005
Dan Clemmensen writes:
> Hal discusses an evolution of hybrids that ends with
> a gasoline-driven motor-generator driving efficient electric
> motors for each wheel. I don't see it. Many of the postulated
> intermediate steps appear to be harder than just building the
> end result. In particular, replacing cam-driven valves with
> electronically-controlled electromechanical valves sounds
> horribly complicated.
Wouldn't it be a natural extension of what we've already seen, though?
Electronic fuel injection replaced mechanically operated carburetors.
Now we are seeing gasoline direct injection systems which largely
replace cam-based intake valve timing with electronic injectors.
Electronic ignition control is replacing mechanical distributors
and rotors. In each case we're seeing a more decentralized design
architecture, with the point of control moving closer to the engine
cylinder. When you look at the complexity of camshafts and timing chains,
it seems to invite a similar redesign. I don't know if such a thing is
really happening, but Huber implies that this change is underway.
> Piston engines are a real mess. Their redeeming virtue is
> high power over a fairly large RPM range. But if you have
> electric motors at the wheels, you don't need this attribute.
> Instead, you build a turbine engine, optimized for maximum fuel
> efficiency at a particular RPM. This engine is a whole lot simpler
> and is well-understood. Diesel-electric locomotives have used this
> trick for over 50 years. A small high-current battery can be used to
> even out the power demand on the turbine. for rapid acceleration.
I don't know if turbine engines can be made to work efficiently at this
smaller size, but the general point seems very accurate: the engine
can be optimized for power generation at whatever RPM and fuel flow
rates work best. When you press on the accelerator, the engine won't
necessarily speed up. You'll draw more power from the battery, and the
engine will run more often and not shut down as much, but it doesn't
have to run harder or faster. It just purrs away, turning gas into
electricity as needed. Such a system should be vastly more efficient
than even today's hybrids.
Huber claims as well that the overall power train will become much
lighter, with no need for heavy drive shafts, differential gears,
transmissions, etc. All you need are cables that run to the wheels,
and motors there which also will do regenerative braking, as hybrids
already do. This will further increase efficiency.
The main problem, I gather, is that the necessary high power density
motors are still very expensive. And of course it is an enormous
redesign of a part of the car that has changed very little in 80 years.
There may be a certain amount of institutional resistance among the car
companies, a need to retrain their engineers and be much more innovative
and entrepreneurial than you would tend to expect from a GM or Ford.
More information about the extropy-chat