'12 Impala 3.6l Police.. Back again... APP P2138 and 5v ref 3 P0697

2 codes present for APP 1/2 correlation (p2138) and 5v ref 3 (p0697).

Pedal has been changed prior to my diagnosis, DORMAN pedal, then warrantied for another, same fault. With factory pedal installed, code returns (only APP code) and reduced power message appears immediately when depressing pedal. Via scan tool, pids show no real drop out or bad correlation through APP 1 and 2 voltage. 5v reference 3 also does not drop out. With scope; back probed into x111 at driver strut tower 5v ref 3 while depressing pedal. Same result, no drop in 5v ref and code returns only for APP correlation MIL Set, and reduced power. Through scan data, both pedal positions are being read by PCM. Removed X1 to check for terminal corrosion and back probe at 5v ref wire at PCM, found some damaged wires to fuel pump supply, and ignition along the way at the harness routing over the trans case. Pulled back loom and tape to investigate more but found no other issues.

Repaired wiring and began to look for my drop again on the scope. Same result. Performed same test with MAP sensor unplugged (the only other sensor on 5v ref 3) and no change. At this point I am convinced we have 3 bad pedals. So I source one from a donor and have it shipped from NC. Same result. So my question is; Looking at all the 4, 5v ref circuits, circuit 3 is the only one unable to truly reach 5v. It stays around 4.96v- 4.98v. (That is the ONLY potential anomaly I see) When all the others are 5v. Is this an issue? I have never had this be an issue, nor do I think it can be. Should I be load testing power and grounds at this PCM then condemning it, or am I so deep down the rabbit hole I am missing something that someone may have experienced in this same issue. After speaking with customer again, they had said that this all started happening after ground cable on neg batt terminal popped off while driving. This is new info, but my data doesn't support a cooked 5v driver or anything similar. However, its a new place to look.  

***Update*** donor factory pedal installed, I have differences in voltages on APP sig 1. Unplugged pedal, I have the same thing at different values. 5v ref code never becomes an issue until vehicle is driven and the fault is recognized several times. Connector X111 is the only inline connector for 5v refs and signals to pedal.

On the scan screen, you can see differences in 5v circuits. And my fault appearing when reduced power mode kicks in. Contact points around x111 have all been opened up and addressed. No faults, pin drag is good, and a wiggle test does not set DTC. Only depressing of the pedal.

Am I having a back feed in voltage at APP sig 1? Why am I seeing voltages with pedal unplugged?

My next step is going to be running my own signal wire from App to PCM sig 1 since wiring is already exposed due to checks.

If you're worried about the 5v ref3 being 4.98v, you can try supplying your own 5v to that circuit and see if the fault goes away.
You can also jump 5v to each app signal to make sure the data reports 5v. This will pretty much prove the signal circuits are intact.

I know it sounds dumb, but I think you may just be dealing a bunch of bad parts. I have had multiple dorman parts for the same car bad out of the box, and the used OEM ones aren't exactly failure proof either.
I'm not sure what you mean in you lastest post by:
"**Update*** donor factory pedal installed, I have differences in voltages on APP sig 1. Unplugged pedal, I have the same thing at different values."
And also:
"Why am I seeing voltages with pedal unplugged?"

The bad parts is where my head is at given the data. fault is only triggered when pedal is depressed.

I have purchased an OE pedal from salvage yard that has same part #.

That is a good suggestion. I can safely jump 5v from the ref 4 wire to the ref 3 wire on the pedal without causing any harm?

To add more clarity to the voltages differences; In the scan data screenshots I provide, you'll see APP 1 and APP2 volts. APP 2 v stays at 0v with pedal unplugged, and .49v plugged in. (its learned value)

APP 1 is slowly declining from .18v with pedal unplugged. Then with the pedal plugged in, it declined from .92 (its learned value) down to .86v.

Why is this delta between APPs 1 and 2 not triggering the fault then? Or even the agree pid that ultimately triggers the correlation code. These are the questions that are making me think its not the pedal, but all points to the pedal.