EPROM emulator

[Tom Hussey]

I understand that the command $05 is download to ram and $06 is download
and execute (OBDI). So you download into ram and then the last frame you
send is download and execute. Too bad my car doesn't support this. Now
it's all OBDII. For sure complete a disassembly to avoid un-plesant
results. For more details about how to use this "bootstrap" see Motorola
Application notes AN1010 and AN1060. You won't want to use the built in
code, just the ideas. Mot and GM worked hand in hand on a lot of stuff.
Seems Mot wrote ap notes on most of it. Another example is the manner
that bank switching is done in '96/7 Camaro. GM implements it *exactly*
as in the ap note.

Cheers,
Tom

Stuart Bunning wrote:
> 
> Sorry if this is a few days behind but I think it is worth mentioning
> anyway.
> It's a bit long (sorry) but I really think this could be the best
> approach
> for some of us anyway.
> 
> As mentioned below use a ALDL command to write to RAM/memcal. Well
> there
> allready is a ALDL command to read from memory so maybe there is a
> undocumented write command that allows to write to memory.
> 
> The factory may have??? used a development memcal with a s-ram chip on
> it
> instead of eprom with a write line coming from somewhere else on the
> ECM ???
> 
> So maybe the micro code is allready there to do such a thing. and a
> custom
> memcal re-write is not even needed. Out of a possible 253 ALDL
> commands I
> have only heard of about 3 that actually do anything.
> 
> As I have been in the service dept of a company that imports
> electronic test
> and measurment equipment for 10 years I have seen plenty of fancy
> calibration devises that work on the standard product with a simple
> bridge
> of two tracks and some serial commands that were otherwise listed as
> not
> used. I have been amazed time and time again of what can be done if
> you know
> how.
> 
> The programmers at GM are not dumb they think of these things when the
> system was origionally designed to aid development and 99% of the time
> these
> (back doors) are never deleted in the production code...
> 
> The big question is has one of this list or diyefi out there allready
> decompiled the ALDL routine section of a 8192 ALDL ECM bin file to see
> if it
> is or isn't implemented.
> 
> EFI DIRECT in AUSTRALIA allready do this with a simple 1 wire
> modification
> and replaced the eprom with a battery backed s-ram on what they call a
> development board. There is also a security device they have installed
> to
> stop people like us copying there technique. But I bet by #$%% that is
> was
> based on what the origional engineer used when the code was
> origionally written.
> 
> Now this may not help every GM EFI tinkerer but just maybe every one
> with
> 8192 ALDL ??
> 
> Think about it.
> if you are serious about his email me off list
> stuart at kenelec.com.au
> 
> >Instead, how about a board with RAM and logic that plugs between a
> >MEMCAL and a P4 ECM - which uses the P4's CPU and ALDL commands to
> write
> >to the RAM.  The MEMCAL doesn't get the processor R/~W signal, but
> I've
> >come up with a jumperless solution using a medium sized EPLD.  Of
> >course, any ALDL based design requires a custom code fragment in the
> >EPROM.  This may reduce the usefulness of such a board.
> >
> >--
> >Ludis Langens                               ludis (at) cruzers (dot)
> com
> >Mac, Fiero, & engine controller goodies:
> http://www.cruzers.com/~ludis/
> >
> >
> 
> Stuart
> Australia
> 1 * 1988 VN Commodore Buick 3.8l V6 (Aussie GM Vehicle 808)
> 1 * 1976 Chrysler Charger 4.3l Hemi 6 with DELCO P4 808
> 1 * Bench ECU for testing and playing 808
> 
> Email: stuart at kenelec.com.au
> ------------------------------------------------------------