O2 input not matching scan data?

[gunslinger]

My scanmaster has been reading my O2 voltage in the mid 500s with the occasional dip low. My fuel pressure was low under boost, so I changed the fuel pump and fixed that, but O2 voltage was still in the 500s even at idle, and the idle was rough.

I hooked up powerlogger, and verified that O2 voltage was actually that low, and was pretty much flat-lined. I installed a new O2, and it's doing the same thing. If I check O2 voltage on pin D7 going into the ECM, it's cycling nicely, with open loop voltage in the high 700s.

All of my fuel trims are screwed up, with most of them around 118-119. Occasionally the scanmaster looks like it's reading properly, because the idle smooths out and O2 voltage goes to around 780mv.
I'm thinking that I've got a failing ECM, but I'm only seeing an issue with the O2 circuit.
Anyone have a good ECM that's already modified for a powerlogger that they want to let me buy? I'd hate to buy a new one, mod it, and have it not fix the issue, or has anyone had a problem with a simple fix?

 

[RmvBfrFlght]

Check the O2 reference ground to the ECM, pin D6. The O2 amplifier in the ECM is differential, with an isolated ground to the engine block.

Another way to put it, connect the DVM between pins D6 and D7 and see what it reads. This is what the ECM is getting. If that is OK, and different from what the ECM is reporting, then likely a bad ECM. Otherwise the ground from D6 to the engine block is bad.

RemoveBeforeFlight

 

[RmvBfrFlght]

Ack, missed that on 1 and 3 wire O2 sensors the body of the sensor is signal ground. This ground needs to be continuous from where it is threaded into the exhaust and the engine block, and then on to the ECM.

RemoveBeforeFlight

 

[gunslinger]

Ack, missed that on 1 and 3 wire O2 sensors the body of the sensor is signal ground. This ground needs to be continuous from where it is threaded into the exhaust and the engine block, and then on to the ECM.

RemoveBeforeFlight

Good point. I'll check for a voltage drop between the sensor body and the ECM grounds. A failing ECM doesn't make sense to me since it is only effecting the O2 circuit.