A challenge for "aldl data stream decoders"

Ross Myers ponty at axis.jeack.com.au
Mon Sep 27 09:40:08 GMT 1999


>I am trying to decode an 8192 baud aldl data stream... (isn't everyone?)
for
>an Aussie V8 P4 (I think).
>
>I connected my laptop to the aldl data stream, went for a drive
>and logged the data described an excel spread sheet: aldl_8192.zip at
>http://www.users.bigpond.com/paul_blackmore/CProgramming.htm
>
>BTW the V8 that is home the the ECU is at:
>http://www.users.bigpond.com/paul_blackmore/VSCommodore.htm
>
>I have "named" some values that I think I can identify but
>I may be wrong. If anyone would like to have a look at the
>spreadsheet and give me their opinion on what any of the
>byte streams (B1 through B56) might represent that would
>be great :).


Oops, Ignore my last message, My fingers typed before my mind caught up.
The Mode 1 data stream address's are located at $0B24 ($8B24 in ECM).
I am basing this off the BLJX6413 VS V8 Memcal not knowing what yours is.
Located from this address is -

0000      =Next message Entry Address
F4        =Message I.D
80        =Use PROM Table
39        =Number of Bytes (57dec)
01BE      =Message Buffer **Can anybody else confirm that?.
01BE      =Message Buffer ** And again??

Then at $0B2D ($8B2D in ECM) the Serial Data message table begins and ends
at $0B9C

The 2 obvious pairs of bytes (#'s 35 and 36) are at $0B71 & $0B73, they
are -

8000    =PROM ID
8001    =PROM ID 2nd byte

This matches your Excel spreadsheet as these 2 bytes never change throughout
your capture and I'm betting your PROM ID must be 072E.

As for the other locations, unless you know the running code you can't
really tell what memory location refers to what (at least I can't!!).
I would try capturing the data again, switch IGN on, don't start and just
move the TPS Up/Down, note the Byte that changes, unplug the IAT, CTS one at
a time and again note which byte changes, and so on.
I did notice that the data stream repeats the location $CE6A many times,
which located at $CE6A is 00, which matches your capture, not sure why
Holden would do that, seems wasteful.

BTW - what software are you using to do all this, I'm very interested :-)

Regards
Ross Myers




More information about the Gmecm mailing list