1227748 refinement level?

[steve ravet]

Nick, there are some places you could look for more info:  Ludis's page
has good info on ECM applications and interchange. 
www.cruzers.com/~ludis.  Also there's a paper on GM idle strategy on the
gmecm page under "papers".  It's for 747 but the control strategy is
likely to be similar, so that may give you some hints also.

--steve

Nick wrote:
> 
> So, I own an 88 fiero 2.5 (TBI DIS) with the 1227748 ECM.  Over
> the last few years I've come to recognize a few traits of the ECM
> which bug me.  Most of them are related to stupid idle tricks it
> plays from time to time.  (I'll explain those at the end of the email..)
> 
> My questions..
> 
> What was the level of refinement back in the 80's on GM ECM's?
> Did "lowly" powertrains such as the 2.5 get ignored in favor of the
> more prestigious platforms?  And thus one could perhaps deduce
> that the PROM code isn't as good as i could be?  Or perhaps the
> whole ECM design?  It's an early P4 design.
> 
> GM used this ECM up until '92 in 2.5-powered vehicles, and I've
> thought about getting the related PROM, but my bottom line (right
> now) is that without the knowledge (and prom burner and
> associated hardware) i am probably asking for trouble.
> 
> Originally I had (don't quote me on these numbers, pulling from
> memory) ACMH9544, which I believe came with the car.  Recently
> I've run ATBS3453 which is the last GM-released PROM specific to
> my application.  No real improvements for the most part, other than
> elimination of an occassional obscenely high cold high idle of ~
> 2500 RPM.
> 
> Is there a "low-brainer" (as opposed to no-brainer) swap to a more
> modern, refined ECM that would have calibrations close enough to
> work with my 2.5 DIS 5-spd without hacking (which i'm not
> currently equipped or knowledgable enough to do?)
> 
> Oh, and for anyone who is curious.  By stupid idle tricks, I mean
> things like.. not being able to "catch" the idle fast enough when
> you suddenly let off the gas, causing the motor to stutter down to
> 500rpm or so before the IAC can catch it and bring it back up to
> 950-1000...Especially when cold.  I'm assuming this is becaues it
> steps the IAC to 0 when TPS > 0, and then steps it back out as
> TPS reapproaches 0, which I don't understand the logic of.
> 
> Not compensating for the additional alternator load when the
> radiator fan kicks in (idle stumbles down to 600rpm before IAC
> grabs it back up) even though the ECM is what triggers the radiator
> fan and it should know better.  At least it's smart enough to
> compensate for the A/C... seems lazy of the programmers to not
> reuse the step-up routine for the radiator fan too.
> 
> The code also will not reduce IAC steps from cold idle choke to hot
> idle unless VSS is 0, and even then it takes 10-15 seconds to fully
> step out the high idle.  Really annoying if it never gets the chance
> to step out the cold high idle before you get on the
> highway...sometimes I get right on the highway and drive for an
> hour, and when i get off the highway and pull up to the light, it will
> STILL be in high idle mode, having not stepped down to normal idle
> yet.  soon as I pull up to the light and come to a complete stop,
> the ECM noticably starts reducing IAC counts and brings the idle
> right down... I wish it would look at the CTS and realise "hey we're
> at operating temperature" and set the IAC to a "last known hot idle
> step count" instead of having to find the idle step count everytime I
> drive it.
> 
> I think a lot of this could be described as just slow IAC response to
> conditions.. and could probably even be corrected with minor mods
> to the idle lag filters.  (I have replaced the IAC with no effect, so it's
> not the IAC itself).  One day I hope to have enough knowledge to
> modify the BIN and tinker/fix it, but as of now I just don't have the
> technical knowhow to relate the assembly code (which i
> understand at face value, i have done assembly before on PC's and
> intel microcontrollers) to what it's actually doing to the ECM and
> hardware and external devices which control the engine.  My main
> problem is code recognition..it's extremely difficult (for me) to look
> at a bunch of ASM commands and figure out... "oh, this is a lag
> filter" or "oh, this is an idle control routine."  I can recognize simple
> stuff (like a table) but it's difficult to figure out just what the table is
> (spark advance or fuel curve or what).  I envy all you reverse-
> engineering gurus!  And yes, I AM reading the 101's....
> 
> nick
> ----------------------------------------------------------------------------
> To unsubscribe from gmecm, send "unsubscribe gmecm" (without the quotes)
> in the body of a message (not the subject) to majordomo at lists.diy-efi.org

-- 
Steve Ravet
steve.ravet at arm.com
ARM,Inc.
www.arm.com
----------------------------------------------------------------------------
To unsubscribe from gmecm, send "unsubscribe gmecm" (without the quotes)
in the body of a message (not the subject) to majordomo at lists.diy-efi.org