[Gmecm] Piggyback ignition modules

Tue Mar 21 13:19:36 UTC 2006

Ok, this logic makes sense.

Here's an alternative to consider which may help you find your solution 
a little faster.  The "7 pin" module used as the "primary" module, 
connected (alone) to the pickup coil, will provide the REF signal 
needed for the ecm and will trigger the ignition coil.  With the bypass 
and EST wires disconnected, spark timing will be controlled by the 
vacuum and mechanical advance mechanisms in the distributor.  The only 
remaining job is to find a way to observe the ECM generated timing on 
the EST line.  This can be done with a 2 channel oscilloscope, which 
can be found on Ebay relatively inexpensively.  Or you could connect 
the EST line directly to the output of your inductive timing light's 
pickup coil.  But if you really want to use the 2nd module, wire only 
the 2 pin power and 4 pin ECM connectors, leaving the pickup coil 
connection open.  Connect power, ground (bolt the module to a grounded 
heat sink), 5V bypass, and EST.  Tie coil- to a second coil.  Loop the 
secondary wire of the extra coil through an open spark plug connected 
to engine ground.  You'll need the voltage generated as the spark jumps 
the gap in the open plug to trigger the timing light.

How is the TC file "incomplete?"  Are the limits not included?  Or are 
they not set up to display correctly?  If you have the REF angle and / 
or one of the limits, you can "read" the definition file using the 
"tuner editor" and obtain the address of the values that are included.

If you have a disassembly of the code you're using, locating the limits 
by hand may not be too difficult.  First, you'll be looking for an 8 
bit value of about 170 decimal or AA hex representing the "base angle." 
  You'll also be looking for a pair of 16 bit values, both of which are 
likely negative (although one may be "zero").  Negative 16 bit values 
start with FF, such as FFE4 and FF39.  You should see a difference of 
about 170-180 decimal between the two limits, which are typically very 
close to the beginning of the calibration data.  You can find 
additional clues by searching through a disassembly for a code pattern 
which looks something like this:

DC 60       LDD     L0060       ;
                        ;
B3 3F F6    SUBD    L3FF6       ;
8D 79       BSR     LAFEB       ; rts delay
                        ;
FD 3F E8    STD     L3FE8       ;
8D 74       BSR     LAFEB       ; rts delay
                        ;
F3 3F DC    ADDD    L3FDC       ;
8D 6F       BSR     LAFEB       ; rts delay
                        ;
93 5E       SUBD    L005E       ;
FD 3F E6        STD     L3FE6       ;
8D 68       BSR     LAFEB       ; rts delay
                        ;
DC 5E       LDD     L005E       ;
FD 3F DC        STD     L3FDC       ;
8D 61       BSR     LAFEB       ; rts delay
                        ;
DC 60       LDD     L0060       ;
FD 3F F6        STD     L3FF6       ;

Notice the repeating BSR (can also be a JSR) after only a line or two 
of code?  If you follow the branch it leads only to an RTS, return from 
subroutine.  This jump/ return cycle creates a delay of a few clock 
pulses.  This is the code area which actually sends timing info to the 
ignition hardware and it represents the "end" of timing calculations.  
If you're using a P4 ecm, expect some of the addresses (0x3FF6, 0x3FDC, 
0x3FE6) to be the same.  Working backward through the code from here, 
look for addresses which reference data in the beginning of the 
calibration data.  A big clue is that the data will be loaded as a 16 
bit value, ldx, ldy, or ldd.  If you check back and forth, referencing 
the data loaded at the location indicated, eventually you should find 
the 16 bit limits.  If you can "speak" assembly, the limit sections are 
fairly consistent in their coding.  You'll find something like:

ldd limitA
cmp timing limitA
BCC (or BGT)
std timing

This loads a limit, checks to see if the timing value is above a limit, 
and replaces the timing value with the limit if the condition is true.  
There's a similar section to make sure timing is not lower than a 
second limit.

As far as the modules, the dual modules connected to the same pickup 
coil may stall unexpectedly if engine speed suddenly drops for some 
reason.  To me, reducing the chances for this are extremely important.  
Creating the potential for trouble often leads to trouble.  Using the 7 
pin module as the solitary trigger and REF signal generator will remove 
this possibility, willl rely on factory electronics for reliability and 
consistency, and will free you to concentrate on solving the timing 
issue.

Just some thoughts.

Zaphod

-----Original Message-----
From: Andrew Gibson <andrewsharyn at yahoo.com>
To: gmecm at diy-efi.org
Sent: Mon, 20 Mar 2006 21:48:46 -0800 (PST)
Subject: Re: [Gmecm] Piggyback ignition modules

   >Do you need the dual modules?  A 7 pin module functions much like 
the
>earlier module without the 5V bypass signal.  And it will supply REF
>pulses for the ecm.

   With my future plans I do need them. I'm using a 1227748 ecm with 
1bbl tbi.
(Runs pretty good so far once it starts)  I need both modules at the 
moment
because I need to sort out the timing tables in my bin file. Tuner Cat 
is the

only one with a definition file and it's not complete as to the max/min 
timing
values. Since this is a dis ecm, my truck definitely has a distributor, 
and I
want est without buring up my engine in the process; I need to hook up 
an extra
coil and timing light to play with different values to see what will 
happen.

   Once runnning the pickup coil can drive both modules. It just doesn't 
have
enough umph when cranking. So somehow I need to amplify the signal, 
split it,
put a resistor inline for the module taking all the signal, or 
something. In a
few weeks when I figure out the timing stuff it shouldn't even be an 
issue, but
I need my truck back and I can't afford to mess up the engine right now.

   Maybe I'll have to wirte a paper up on this when done so everyone can 
see a
new option for the plentiful iron duke tbi setup. Hey if enough of us 
call one
adapter company they might be willing to make a generic adaper to fit 
all the
1bbl intakes for gm, ford and jeep.

---------------------------------
 Yahoo! Mail
 Use Photomail to share photos without annoying attachments.
_______________________________________________
Gmecm mailing list
Gmecm at diy-efi.org
http://lists.diy-efi.org/mailman/listinfo/gmecm

___________________________________________________
Try the New Netscape Mail Today!
Virtually Spam-Free | More Storage | Import Your Contact List
http://mail.netscape.com