Re: Continuous IJ

[Scott Fisher <sefisher@cisco.com>]

G. Benedikt Rochow writes:

> Call me ignorant (which I am in these matters), 

Don't worry.  Being ignorant is easily fixed; being *stubborn* about it
is where people run into trouble... :-)

> but somehow, that sounds to me a lot like a carburetted engine:
> mix fuel and air, suck mix into combustion chamber.

What you've just described is an internal combustion engine, period.
Emulsifying fuel and air is key to any spark-ignition engine, whether
you emulsify it in the carb throat or in the top of the cylinder head.

The main thing FI buys you is precision -- the fuel delivery system can
meter much more precisely than most carbs can, and this is important in
meeting modern emissions regulations.  This precision and its associated
controllability mean that an FI system can be modified under the control
of the oxygen sensor, something that would be very difficult to do with
a carb.  (Not an SU carb, of course, but may well be the only person on
this list who likes them; like Unix, SU carbs are expert-friendly. :-)

Also, fuel atomization in the combustion chamber can make a big
difference in performance/economy/emissions, depending on a LOT of
factors.  I would guess -- and this is purely an educated guess -- that
a lot of work went into studying how the fuel was atomized as it got
drawn through the partially-opened intake valve, as the piston drops on
the intake charge.  

> (Do the injector really squirt fuel all the time, without
> interuption?

Yes, in CIS systems -- that's what the C stands for, continuous.

> If so, what's the fuel dist. for?)

Distributing the proper amount of fuel to each intake port.  

There are also *timed* or *sequential* fuel injection systems, which do
what you (implicitly) seem to think fuel injection "should" do -- pump a
measured amount of gas into the port that's currently in operation, and
not any other till it's that port's turn.  Alfa's SPICA fuel injection
system worked this way, for example -- which makes sense if you
understand that it was derived from a diesel injection pump (and if you
understand about how compression-ignition works).  SPICA includes a
number of complex mechanisms to keep the injection in proper
synchronization with the engine, to compensate for RPM and load, and
other concerns (thermal adjustments, for example).  It was a mechanical,
Sixties-era solution to the problems that, in the Nineties, are handled
with air-fuel maps, electronic sensors, and EPROMS.

I have been led to believe that the mechanical fuel injection on the
Mercedes 300SL of the Fifties was also sequential, rather than
continuous, but I'm not certain and I'd love to hear an authoritative
response.  And likewise, I seem to recall that the famed Kugelfischer
injection systems of the early Seventies was also sequential in
operation, but as it's typically only on cars that I haven't yet been
able to afford, I remain ignorant of its details (but not stubbornly
so).

--Scott Fisher