Re: 7749 guru needed....

["Green" <green@dsl.telocity.com>]

>
> Ummm, it sounds like you are using a 3-bar sensor without the software
> patch to enable use of a 3-bar sensor!
>

He used my program for base, so it is for 3 bar (which i hope is done
correctly).  Runs fine in stock cases.

> The patch ought to do the following:
>   1) An  ASLD  instruction (aka "times two") in the fuel code needs to
>      be changed to do a "times three" instead.

Instead of fighting for code space to do this and/or recompiling (as it's a
bit tight for code in the stock location this is done), I used another
fueling table (BPC value @ F28, closest thing to an injector constant) and
multiplied by 1.5 there, giving same effect.

>   2) Several parts of the function at $F4F0 need to be changed.  This
>      function computes the "fake" 1 bar MAP.  It also calculates the
>      1-bar and 2-bar table indices.  Finally it adjusts the MAP value
>      to a true absolute reading (sensor 0 does not equal hard vacuum,
>      the SpeedDensity formula needs a value where that is true.)

Changed that as well;  there's a calc that computes the fake 1 bar and
checks for overflow. Basically had to get new constants in that calc to take
~ 1/3 of full range for vac, rather than ~1/2 from the 2 bar.  I believe
this is the only part of that code section that needs changing ?

>   3) There is likely an initilization code fragment that duplicates a
>      small portion of the $F4F0 function.  This fragment also needs to
>      be changed.

Haven't seen this (or at least haven't changed it).

So the big code change i made was the 3 bar to vacuum calc.  The changes in
the tables portion of the chip are extensive also.  Tables using full range
MAP have to be compressed & then extrapolate values for 15psi to 30psi,
values using full range MAP have to be changed.  Vacuum constants & tables
stay the same (even though the vacuum calc is a *tad* off; i may change that
so it's accurate).  Had to go thru to make sure to see what calcs were using
what values (full range map, boost, vacuum/fake 1 bar etc).  After doing it
this way, some things become a bit difficult; like VE table tuning (was
difficult on 2 bar too actually).  The 2 tables (open throttle and closed)
now have a bit less resolution; not a big deal at idle, but at open
throttle, you only have like 2-3 lines of table in vacuum, for drivability
tuning.  The closed throttle table wastes a 6x9 table space of low RPM from
vacuum to 3 bar of boost !  I'll probably not see 1000rpm and 28psi  :-P
So i'm hoping to change the way the VE tables are setup now.  Maybe make one
for drivability, that uses vacuum, then make the other in use during PE mode
only, using full range ..? I'm not sure yet.

Any info/advice is appreciated  :-)

Brian Green
91 GMC Syclone
12.06@113 and hoping for 10s this year



----------------------------------------------------------------------------
To unsubscribe from gmecm, send "unsubscribe gmecm" (without the quotes)
in the body of a message (not the subject) to majordomo@lists.diy-efi.org