[SGVLUG] [OT] Especially for Tom & his Prius.... [my rebuttal, then I'll shut up]

[Dustin Laurence]

On Wed, Jul 19, 2006 at 02:19:00AM -0700, David Lawyer wrote:
> 
> On Sun, Jul 16, 2006 at 09:58:53AM -0700, Dustin Laurence wrote:

> Part of the reason was that I didn't read the thread from the start
> but picked items in the middle.

I realize you're reading this with some delay because you're on dial-up
(or at least I suspect that is the reason), but it often leads to a
inadvertent reboot of the thread where you revisit issues we've finished
with.  It gets difficult to tell if you're changing the subject on
purpose, missed context, or are responding to the subject and context
from earlier in the thread.

> equally efficient.  They are not.  But I previously stated that a
> diesel was not too much different in efficiency if you compare it to a
> gasoline engine at full throttle.  I'll try to look more into this as
> I could be wrong.  But I've got data that I just looked at that may
> imply I'm right.  More about this later.

It's an interesting point, albeit not relevant to vehicles until
practical hybrid drives existed.

> > If you want a fair comparison, you may compare a diesel with 40% better
> > mpg than the gasoline engine.
> 
> efficiency would be the same in this case.  Right now I would like to
> know for sure what the maximum thermal efficiency of an automotive
> diesel is.  I know that railroad locomotive diesels approach 40%
> thermal efficiency yet go significantly lower at off peak loads.

Marine diesels and fixed generators are probably even better, since size
matters.  Big cylinders lose proportionally less heat through the
cylinder walls, for one thing, and I suspect the low RPM operation helps
as well.

This engine

http://www.bath.ac.uk/~ccsshb/12cyl/

claims >50% thermodynamic efficiency for a 1.5 million cubic-inch,
100,000 HP engine (but it's the torque figure that truly impresses at
5.6 mega-foot-pounds).  Irrelevant for anything but large ships, of
course, but fun to think about.

> should get say 35% more miles per gallon.  But the info I have (and it
> could be wrong) seems to imply that peak efficiencies aren't that much
> different.  The peak efficiency of diesel should be higher due to both
> it's higher compression ratio and faster burning.

Higher compression ratio, yes, but the fuel burns slower because of the
stratified charge and the lower flame speed.  This is why diesels suck
at high-RPM >~5000 RPM.  Also the fact that a diesel can run leaner than
SI, in fact always runs lean.

> ...But per tests in
> 1949, a gasoline engine (installed in an Oldsmobile with water pump,

[[snippage]]

> Then I've got a personal communication from Daimler Benz (Mercedes)
> dated 1975 in Germany showing a peak diesel thermal efficiency of only
> 27 1/2 %.  This is with road load at 70 mi/hr.  It could be higher say

I'm afraid that all these are two old to be very relevant, as you
suggested.  Diesels have evolved more over the intervening years than
spark-ignition engines, and on the other hand there are advanced SI
engines that shut off cylinders to reduce pumping losses.  Really things
are moving quite fast and you probably need data within the last ten
years or less to make useful predictions about the near future.

Your argument about efficiency at peak RPMs is most interesting if we
have a practical hybrid drive mechanism of some kind to store energy,
whether as electricity or something else (there are at least concepts to
store mechanically with compressed air or flywheels, possibly not
practical but it doesn't matter for this argument).  Otherwise we can't
just run the engine at it's optimal load and RPM.

I need to stipulate that you've already argued about your driving
method--you're wrong, it's neither practical nor does it do what you
think, I've told you why, and I will not discuss it further because it's
boring and repetitive.  So I am *only* talking about storing energy
internally and *not* in the car's own bulk kinetic energy.

So now talking theoretically, the most efficient heat engines always
have the greatest ratio between reservoir temperatures.  The most
practical way to do this is probably running a combined-cycle engine
with a turbine, since then we can run very high hot reservoir
temperatures and use the exhaust to drive another heat engine (and
thereby lowering the effective cold reservoir temperature below what a
turbine can normally do).  A Stirling engine is nice for the second
stage because it is simple and closed, but it's possible that an
automobile could use a steam second-stage like fixed generators already
do (and I think I recall a major automotive company playing seriously
with this, but I don't recall who nor do I feel like hunting it down).
People have a hard time these days thinking of steam and cars in the
same sentence, but it can be useful in a high-tech context as well as
low--in any case the game is just to run some sort of external
combustion engine on the exhaust of the hottest first stage you can
efficiently run, and to figure out how to do this without adding too
much weight.

As a total aside, hybrid drive trains very much give Stirling engines a
potential new lease on life, since they are practically the poster child
for engines that cannot be throttled and want to do just the one thing
they do at the speed and power output they wish to do it at.

Restricting ourselves to reciprocating engines, I think it's very hard
to say what the ultimate efficiency will be or how SI and CI engines
will compare since there are so many places that still need development
on both types of engines.  Eliminating the cam shafts is a huge one, and
that may benefit SI engines more (not sure about that), and of course
there are various interesting combustion concepts in varying states of
fantasy, some of which blur the line between the cycles somewhat.  And
the relevant "efficiency" is in the context of the whole automobile,
which actually favors SI a bit--the greater weight of CI engines isn't
great, especially in small light cars where fuel efficiency is most
likely to be a dominant consideration.  If we ever care about fuel
consumption enough to start using space frames and composites, that
extra engine weight could be a pretty significant percentage of the
total vehicle weight.

That also actually works against both hybrids and combined-cycle
engines, of course, since they involve and even larger additional
drivetrain weight penalty than diesels.  Hard to say which factors
ultimately win.

With what we know how to do now, both types of engines have pushed ahead
in different areas and that technology will cross over; SI engines are
going direct-injection, for example, and CI engines will probably go
toward multiple cams.  It's hard to say how that will shake out but I
will guess that CI will retain it's efficiency advantage for the time
being.  Not only is the compression ratio higher, but CI turbocharges
much better than SI (and high-octane fuel is only going to get harder to
obtain), and a turbocharger is one of the few things that genuinely
increases the thermodynamic efficiency of a reciprocating engine by
reclaiming waste exhaust heat (it's sort of part way to combined cycle).
Some advanced SI engines are already reducing pumping losses by shutting
off cylinders (like Cadillac did years ago with the 8-6-4 engine, but
this time it really works), which doesn't seem to have closed the gap so
I don't think the greater efficiency of diesel is due to pumping losses
alone (but I don't have detailed information on that.

ULSD in the US is sort of critial, because if you can't sell advanced
diesels in California and New York the major automakers of the world
just aren't going to put the effort into diesel they put into gasoline.
Unless investments are similar the technology that gets the most
investment is going to be the winner.  Mercedes and VW seem to be taking
the lead here since their home market is so diesel friendly.  If/when VW
overcomes it's irrational dislike of hybrids and makes a turbodiesel
hybrid it will be interesting to see what happens.  I fear they will be
scooped by someone less buttheaded and more visionary, like Honda.
Well, it was their game to lose, so if they lose it they have no one but
themselves to blame.

Dustin

-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 189 bytes
Desc: not available
Url : http://www.sgvlug.net/pipermail/sgvlug/attachments/20060719/5332fcb5/attachment-0001.bin