Re: SPICA, then ramblings

[James Montebello <jamesm@domain.elided>]

On Thu, 21 Nov 2002, Mike Nakamura wrote:

> Did ALFA copy a diesel injection system for SPICA?

No.  SPICA modified their own diesel pump for use with gasoline.  This was
common practice at the time.  The two of the other commonly available
mechanical gasoline injection systems (Bosch MFI, used on Porsches and
Mercedes; and Kugelfischer, used by BMW [and Porsche, on racing cars])
were created the same way.  The origins of the Lucas system are obscure
to me, although I believe it was a licensed copy of the Bosch system.

All of the designs are broadly similar.  It's been said by someone who
runs a business rebuilding the Bosch and Kugelfischer pumps that the
SPICA pump had the best detail design of the lot.


> It's my view that mechanical injection was over taken by electronic
> injection just as carburetors were superseded by electronic injection
> because electronic injection was cheaper to tune for the best compromise
> between low emissions and drivability. During the push for emission
> control, electronic controls were incorporated into (or hacked onto)
> carburetor systems. If electronic injection was not available,
> mechanical injection may have prevailed.

Electronic injection replaced mechanical injection primarily because it's
vastly cheaper to produce, and more reliable long-term.  The injectors
are simple, cheap devices.  The electronics don't wear out and change
their settings over time.  Other than watching out for age cracks
in rubber and plastic fittings, and the usual corrosion issues for
all automotive electrical fittings, EFI systems are reliable enough
to be ignored for years.  Digital electronics also brings with it the
ability to have fuel and spark control algorithms that are difficult to
impossible to perform in an all-analog design.  One basic EFI system can
be designed for a whole range of cars, with only the maps locked up in
the ECU differing from model to model.

Meanwhile, mechanical systems wear as a function of normal use, and
eventually need to be rebuilt.  They require regular maintenance, and
they are prone to neglect, abuse, and misadjustment.  A particular
pump design is fairly specific to an engine model.  The fuel cam
could today be easily designed by a computer program, the prototype
made out of the 3D CNC systems.  In the 70s, however, this was a
laborious process that no doubt took weeks.


> I wonder if US emission and mileage standards brought hi-performance
> electronic engine management systems to us faster or if there would have
> been market demand for them. The computer has affected automobile to
> such an extent, it's daunting to begin thinking about the impact.

It started with the EPA's demands, esp. in the case of the US automakers,
who were quite intractable on the issue.  The early designs on most cars
were awful, and badly tuned, and an entire generation of car buyers
came to curse FI primarily because of the baggage that came with them
(thermal reactors, bad 2-way catalyst designs, a zillion vacuum hoses).
From roughly 1975 to 1980, emissions systems on nearly all cars were a
liability, and EFI was unfairly lumped in with them.  Once the engineers
figured out how to get relability from the systems, however, they began
to use the new flexibility EFI gives to have power AND good emissions
AND good economy.  The public responded to the added power and economy
favorably, so now it's the market driving improvements.  The EPA is
squeezing again (0.01% of the emissions of a 1970 car isn't good enough,
it has to be zero), so people like Erik Storhok stay in business, trying
to balance things.

james montebello
--
to be removed from alfa, see /bin/digest-subs.cgi
or email "unsubscribe alfa" to majordomo@domain.elided