Six_Shooter

iTrader: (2)

This is something that I have wondered about for a LONG time, and have not been able to find a good explaination.

How is it that different bins that use different masks (XDF, ECU,etc), all work on the same ECM? Said a little differently, the Same BIN file used on different ECMs.

Examples:

$8D, $8F, $88, $58, $59 all work on '7749 and '7730 (other's as well).

The data or tables are at different locations between many of these bins, and use different XDFs to adjust them.

$8F, $A1 works in '7730 and '7727, again data and tables are at different locations with in the bin.

Is there something within the bin file that points to certain locations when looking up tables?

Back when I was first looking into tuning, it was said that the ECM had the "mask" programmed into it and is why you had to match the bin file to the ECM to have it work, this doesn't seem to be the case, with all this nin swapping between different ECMs and ECM swapping with the same bins.

TIA

JP86SS

iTrader: (1)

The main issue is matching the ECM wiring to the mask.
Hardware outputs in software are mapped out in the configuration registers to use them correctly for each mask. From there the software is programmed to operate the input/output reletive to the operation desired.
Some small "shifting" of positions between masks does exist due to using the outputs for a function that didn't exist or wasn't used in some applications.
A quick check of the wiring would show any major concerns to check into.
Several of the functions that are different have software "control" bits to set if the output is used for one function or another.
As far as the XDF and ADS, Those items can be manipulated any way needed within the software.
Table locations can be anywhere within the addressing range. "Masks" keep the locations consistent on the editable parameters so making changes could be done without rewritting the actual software.
Data logging output pulls data from any active memory location (RAM) and stores it to be output sequentially. The data item can be anything in active memory.
The routine for outputting is set up in software (to make the hardware do the transfer to the output)
That routine seems to be setup for the maximum standard output length that can be performed within the cycle time available. There is a bit of extra time to add additional data items but I'm not sure how much.
HTH

RBob

iTrader: (1)

I think this is where the confusion comes from. Not all GM ECMs are set up this way. The '7730/'7727/'7165 are ones that are not. When the PROM is removed ALL code and ALL calibration values are also removed. Everything is in the PROM.

With these ECMs it is like being able to run Linux or Windows on a PC.

Then there are the ECMs such as the '8746, '6870, and the '7302 ('89-'92 TBI, '85 TPI, '86-'89 2.8l MPFI). These ECMs have the majority of the code in on-board masked ROM. The PROM contains all of the calibration and a little of the code.

With these ECMs the mask has to match up with the ECM Service number. Otherwise the code is going to the wrong locations for the various calibration parameters.

RBob.

Six_Shooter

iTrader: (2)

That makes sense.

So is it safe to assume that the C3 ECMs have the code on board and the P4 are capable of swapping to different code?