Turbo control in a GM ECM

sfg sfgent at ac.net
Sat Oct 4 21:33:43 GMT 1997


Ludis Langens wrote:
> 
> I've been lurking on the list here for several months.  If I find the time,
> I'll have some comments to add to the "hijacking an OEM ECM" thread.
> 
> Anyway, I'm currently figuring out the code from a GM ECM.  It is of the older
> C3 generation.  This ECM has support for a turbo or supercharger.  All you
> need to do is flip a bit in the EPROM, drop in new calibration tables, change
> the MAP sensor, plumb up a turbo, and you can be a road terror.
> 
> I've concentrated my efforts at the newer (second) version of the code.  This
> version has a few flags and variables that seemed to be meant to interface to
> a calibration box.  I recently disassembled the first version of the code
> (used only in 1985).  The extra variables are used by a long subroutine that
> is missing in later years.  This subroutine has something to do with turbo
> control.  It controls a PWM ECM output and checks an extra on/off ECM input.
> 
> Only one vehicle/engine combo ever used this ECM.  Because it was never sold
> with a turbo, I can't consult factory wiring or repair manuals to see what
> the output and input connect to.  What sort of electronically controlled items
> might be used in a turbo setup?  During periods of no boost, the PWM output
> is turned fully on.  When there is an engine fault, or when the on/off input
> is grounded, the PWM output is shut off.  Otherwise, during boost, the PWM
> output is varied based on the amount of boost and the throttle position.  I
> haven't yet followed through all of the subroutine's logic.  It would help
> if I knew what the output controlled.
> 
> The MAT sensor polling function is one place that interfaces to the turbo
> subroutine.  Whenever there is or recently was turbo boost, this function
> does not check the MAT sensor readings against the normal hot and cold
> (shorted/open) limits.  In the newer code version, the function still has
> vestiges of this logic, though it can never be triggered.  Under turbo boost,
> is it reasonable to allow very hot and cold MAT readings?
> 
> A related question:  Both versions of this ECM contain an unused block of
> code that has something to do with the O2 sensor.  An option bit in the EPROM
> selects the "normal" code or the "special" code.  None of the calibrations
> used in this ECM select the "special" version.  The "special" code is 234
> lines and replaces 105 lines of the "normal" code.
> 
> One significant feature of the optional version is that it controls how fast
> the O2 sensor filter responds.  Note that this vehicle/engine uses a single
> wire (unheated) O2 sensor.  Would a heated sensor need special handling?
> Is there something else that would require using the O2 sensor differently?
> 
> BTW, I'm not interested in hooking up my own turbo.  I just want to understand
> everything that this ECM does.  Besides, a turbo is already available in the
> aftermarket.
> 
>                    unsigned long BinToBCD(unsigned long i) {unsigned long t;
> Ludis Langens         return i ? (t = BinToBCD(i >> 1), (t << 1) + (i & 1) +
> ludis at netcom.com                 (t + 858993459 >> 2 & 572662306) * 3) : 0;}

Ludis,
The only thing that comes to mind that is left to control might be a
waste gate. The little Chrysler Turbo Colt in '86 had an ECU controlled
waste gate as I recall. My guess is that the ECU you have under study 
might have routines similar to a mid 80s 350 turbo Trans-Am. I don't
know if this car had a wastegate or not. 

Posts like yours is why I subscribe to this list. Not radar detectors.
Bravo. Please keep us all apprised of your progress in disassembling
this code. Steve



More information about the Diy_efi mailing list