Modifying OBD-II systems

EK sk8er at netmost.com
Sat May 15 14:04:54 GMT 1999


-----Original Message-----
From: Bill Shaw <bshaw at connix.com>
To: diy_efi at efi332.eng.ohio-state.edu <diy_efi at efi332.eng.ohio-state.edu>
Date: Wednesday, May 12, 1999 1:28 PM
Subject: Modifying OBD-II systems


I have had mixed results with a "OBD1 compliant" flashed based VCM used in a
1995 S-10 with a "W" CPI V6.  Standard scan tool ALDL does not appear to
work with it. I have tried one of the new B&B electronics AutoTap units, and
it will not talk at all. The Autotap does work on newer GM stuff. I talked
to one of the guys at B&B, and they are supposed to have new software out
for the AutoTap within a couple months that will support my particular VCM.
I have also talked to the guys at ACCEL about their powertap, and that
device is not compatible with my unit, nor are they interested in making one
that is. I pried the VCM case open to take a quick look, but did not have
time to take any useful data. Remember seeing 2 separate PCBs, 2 wide SO
flash -looking devices, and a 132 lead pqfp processor, with Motorola type
markings- guessing a 68332 derivative.
I was able to successfully fake out the cat O2 sensor by sending the VCM a
constant bias voltage of .766V For actual operating monitoring, I hooked the
2 O2 sensors to a 2 channel DSO scope, and monitored the cat 02 changes,
from .68V to .9V operating in ranges of idle to WFO. Idle voltage stayed
very close to .70V. The fake Bias voltage is picked off from TPS reference
voltage and divided down with a 10 turn 50K pot to .766V. I connected the
ground from the oxy sensor to VCM to a local ground, as I believe the oxy
sensor ground is used to cancel any thermoelectric voltage, which you
probably don't want to add in this case. This setup has been OK with no MIL
for about 9 months thru both hot summer and mild winter temps. Have had
other problems recently, believe Sticking EGR that turned on MIL, but
lightly "tapping" EGR valve turned MIL off after 3 driving cycles. GM Shop
manual is very helpful, stating typical limits before trouble codes set. At
this point in time, at least for me, I am not sure if it is more practical
to fake out sensors, replace VCM with easily hackable computer, or wait for
someone to come up with a reasonable cost OBD flash reprogramming tool.

Eric







More information about the Diy_efi mailing list