Bench ecms

[Lawrence King]

Hi:

Ideally for any measurement system there is just one point which is
defined to be ground. All sensors should have seperate wires which
lead from this single point (called the star ground). The star ground
should be really close to the ECM (in fact it would be even better if
the star ground was inside the ECM). The same applies to the +5V. Do
not share powers and grounds between sensors. Do not use the frame as
a return path.

The frame may have only 0.010 ohms resistance, but when you crank the
starter it draws hundreds of amps. V=IR in this case R=0.010 and
I=100, hence the voltage drop (V) across the frame is 1 volt, this is
a very significant voltage to your sensors (ie 20% error). Go to the
effort of sopplying seperate pairs (or triples) of wires to each
sensor, it will save you a lot of grief.

		-Lawrence-

Bruce Plecan wrote:
> 
> I had been told the sensor grounds on a gm ecm floated in reference to the
> chassis ground, well that doesn't appear to be correct, everything is just a
> couple tenths of an ohm from true ground.
> With no displayed voltage drops across grounds.
> While still seperated, does it really matter if sensor grounds go to
> chassis, on an actual car?.    If so then where Have I strayed to having
> had sensor ground problems????....
> 
> For coolant temp, and intake air temps, I was just going to share
> resistances.  Tieing grounds together.
> 
> For TPS, and MAP I was going to tie the grounds together, use the
> +5 from one ecm, and then share the sensor's output.
> YES/NO??????..
> 
> Cheers
> Bruce        RPM signals applied to both ecms, and powered up
>                   leds flashed and no problems today.
>                   The cat has grown boarded with the flashing LEDs,
>                   but the staff is acting like their inna stupor.

-- 
Lawrence King   lawrence at promobility.net    Ottawa Ontario Canada
70 Buick Wildcat,        71 Lotus Elan,          92 Nissan NX2000
I am a three car family (four if you count the Buick as two cars)