Accel enrichment

[Joe Boucher]

Okay you scurvy dogs, why don't you sparkies and software geeks  move
aside and let this M.E. type have some elbow room and REALLY make a fool
of himself and try on a HUGE cone shaped hat.

Once apon a time, during my senior year, I would bury myself in the
deepest recesses of the library during the day to study.  When I became
bored or tired (about 47 seconds) I would refreshen myself by wondering
around and finding interesting books.  I found a set of automotive engine
text books.  One book was devoted to carburators.  Full of differential
equations.  One of them was for the flow rate through the carb.  It was
about 3 lines long complete with many partial differentials of many
different variables.  Might as well have been greek.  At steady state,
all the partial variables went to zero and the equation left was nice and
short.  Just my kind.  So, what's the point?  During acceleration, the
flow rate is non-linear.

This was confirmed by some lazy Saturday afternoon channel flipping.  The
show had a late 70's Nova hooked up with an accelerometer and vacuum
gage.  The Nova did a "jack rabbit start".  Big drop in vacuum.  No
surprise. They  then did a slower, much smoother acceleration.  Even
though the acceleration was much slower, maybe a fourth, the drop in
vacuum was disproportionatly large.  It was 3/4th the gage value drop of
the rabbit start.  I have confirmed this with the vacuum gage in my
Camaro.  Small increases in throttle opening cause big drops in vacuum
during the acceleration to the new steady state value.

Let's review.  Steady state at 20% throttle equals 19 inches of vacuum.
40% throttle equals 16 inches.  When the throttle snaps from 20% to 40%,
the needle will drop to as low as 10 inches.  The pressure in the
manifold goes beyond the upcoming steady state value, then sinks back to
the new value.  That means there are more molecules bouncing around in
the manifold than at steady state.  I can't for the life of me visualize
why.  But consider this.  There is a column of air tens of thosands of
feet high trying to cram more air molecules into the manifold and the
throttle is restricting that flow.  Nature abhors a vacuum.  The air in
the manifold may have inertia and resisting a flow rate change, but that
big, gravity fueled column of air isn't lacking in push either.  For you
sparkies, an anologous example would be hooking a transistor radio to a
nuclear power plant generator.  Lot's of overkill.  No shortage of
electrons or molecules.  On a Chevy 350, the reason for all this mayhem
is  4" wide piston moving downward quickly while trying to suck air
through a, at best, 2" valve opening with a big piece of metal stuck in
the middle of the air flow.

So what's my point?  Well, the fuel may be flashing out of solution as a
previous poster suggested (I thought the amount of liquid suspension went
UP when pressure increases, that's why it rains when the temperature and
barometric pressure go down) but, for some mysterious reason, there is
also more air in the manifold and therefore to keep the proper recipe,
more fuel is required for acceleration.

Joe Boucher
'70 RS/SS Camaro  '81 TBI Suburban

Tony Bryant wrote:

> Why is accel enirchment (== throttle pump) needed?
>
> I'm talking about the momentary increase in required injector
> duration when the throttle is quickly opened, as opposed to constant
> running at WOT at that RPM.
>
> For example a transition between from 15" vac to 0" (WOT) at say 2000
> RPM.
>
> I'm trying to optimize this part of my algorithm, and I need to
> have some sort of understanding of what I'm trying to fix.
>
> What is going on in terms of MAP, air flow velocity, pressure,
> fuel vaporisation and VE?
>
> What algorithm are in use, and what do they depend on?
>
> Any help greatly appreciated.