[SGVLUG] [OT]Hybrids and trains (was fuel prices and the dollar)

[David Lawyer]

On Wed, May 14, 2008 at 10:30:14AM -0700, Mike Rubel wrote:
> I know this isn't Linux-related, but I have to take issue with a few of
> these points.  Others have already written in, but I like to hear myself
> talk, so...  :)
> 
> Disclaimer: I don't own a hybrid, but I do commute by train and bicycle.
> 
> > Hybrids use internal combustion engines no more energy-efficient than
> > the engines on non-hybrids.
> 
> I fear this is a misleading statement.  The engines may or may not have a
> higher Carnot efficiency, but if they are used more efficiently, then they
> can consume less fuel per mile, and that is what happens in practice.
Correct, but only because other small cars are not designed for
extensive coasting, etc.
> 
> > They also waste a lot of energy converting the gasoline motor power to
> > electricity, charging a battery, and then withdrawing energy from the
> > battery to power an electric motor to move the car.
> 
> This isn't how most hybrids work; instead, they have multiple energy
> pathways.  In the case of the Prius, there is a purely mechanical linkage
> from the engine to the drive wheels, and most of the power goes over that
> linkage.  There is another, more lossy, pathway via the motor/generator
> and batteries, which is used to smooth out power delivery during
> variable-speed driving.  It comes at a cost in weight and complexity, but
> obviously that cost is offset by other gains, or they wouldn't bother.

True, sort of.  But if pulse and glide were used as well as cars
designed for coating, the extra weight and complexity would not be
worth it.
> 
> > All this energy conversion wastes energy and adds to the
> > weight of the car.  The reason hybrids get good mileage is not because
> > they are inherently efficient, but because people don't know how to
> > efficiently drive a non-hybrid and because laws and car design, etc.,
> > impede one from efficiently driving it.
> 
> Yes, all else being equal, a non-hybrid could be driven more efficiently
> under approximately constant-speed conditions, due to its lower weight. 
> Under real driving conditions, however, hybrids do better.
I was thinking of real driving conditions with a non-hybrid doing a
lot of coasting, and pulse and glide.

> 
> > Efficient driving will mean doing a lot of coasting but coasting in
> > neutral is illegal.
> 

> In which states is coasting in neutral illegal?  When driving a manual, I
> coast when possible, especially down long hills.  I never realized I was
> breaking the law.
I think it's still illegal in California but I could be wrong.
> 
> > As for trains vs. airplanes, it turns out that they are about equally
> > energy efficient, and we don't save energy by taking the train.
> 
> That's simply too strong a statement, it depends on a lot of variables,
> including the fraction of filled seats, the number of stops, the length of
> the trip, and the efficiency of the equipment.  It may be that Amtrak
> trains, which often run with only a small fraction of seats filled and are
> typically used for short trips, have comparable per-passenger fuel
> consumption to airplanes, which are almost always packed and are used for
> longer trips.  It does not follow that "we don't save energy by taking the
> train" in general.

Trains inherently have lower load factors (% of seats full) since they
make a lot of stops.  If a train has 100% of seats full between two
stops, it may have well under 100% of seats full between other stops.
But I think that Amtrak does have fairly good load factors but not as
good as airlines, partly for the reasons stated.

There are likely some situations where the train is more
energy-efficient so I probably should have said "we don't always save ..."

			David Lawyer