Persistent P0138 Code (Cherokee content)

My Cherokee has had a P0138 code pretty much from the day it was up and running and on the road back in October. I've been dicking around on getting everything on it fixed for inspection for too long now (and driving around with no inspection stickers in the mean time) and the CEL is the last hurdle to clear. Problem is, I'm at the point where I don't know what else to try. So this is where you come in, good folks of GRM.

Relevant details:

-The rear O2 sensor wires were damaged right before the plug (chassis side). One was severed, another had the insulation damaged, the remaining two were untouched. There was a plastic clip that held the plug to one of the sections of floor pan I replaced and I forgot to make other provisions for keeping the wires away from the driveshaft. I suspect the hanging wires caught on the u-joint, ripped the wires off right at the sensor, and damaged the wires further upstream near the plug. This likely occurred on my first actual drive of the car, which is also when the first CEL appeared. However, because the battery was totally dead prior to that point and the Jeep hadn't seen regular use in over a year, there may have been an additional pre existing condition that would have also triggered a CEL.

-My initial repair on the wiring didn't have the best solder joint. The wire was broken very close to the plug and I didn't have much to work with. This was later rectified (see below.) HOWEVER, the CEL was cleared and stayed off for a while after this.

-O2 sensor was replaced because of the aforementioned sheared off wiring. New NGK direct fit part.

-Jeep first went in for inspection around this time, and failed for a "rust hole" in the cat (which was kinda bogus, but I digress...) and a pending CEL relating to the other 02 sensor, which I was unaware of. Forget the code, but it was different, maybe a P0142?

-Upstream O2 sensor, cat, and exhaust were then replaced. Again, brand new parts including a direct fit NGK sensor. Code was cleared. Everything is hunkey Dorey for a few days. Then...

-CEL reappears, P0138 again.

-Apparently there's a fuse for the 02 sensor circuit, so i check the fuse. Fuse appears fine, although I didn't actually replace it.

-So I decide to redo my wiring harness fix. Pulled the plug itself totally apart and soldered in a patch section of wire. Repair is now much more satisfactory. Clear CEL, then driving to dinner later that night...

-CEL returns after about 15 miles. berkeley me!

And that's where it has stood for the last two weeks. The light did go off for half my morning commute one day last week, only to return before I left the parking lot on the way home.

So what now? I feel like the wiring issue is pretty much the common denominator all along, but I'm really not sure what else I can try there. Are the solder joints in the wiring causing additional resistance in the circuit? Or am I chasing a red herring here? What can I try as far as diagnostics? I have a code reader that can suppossedly do data streaming stuff, anything helpful that that can provide? Do I just move to the next county, where I don't have to pass emmissions?

I'm really kinda stumped here and I've had more than enough of firing the parts cannon at this hooptie already.

TL/DR: WTF is wrong with my Jeep?

Make sure you have heater power for the O2 sensor. If all else fails try a MIL eliminator to pass inspection.

I have been told that soldering O2 sensor wires is a NO-NO. Reasons given were the increase in resistance and the fact that some O2 sensors draw an external air reference thru the wires. Googling "soldering O2 sensor wires" brings up numerous mentions.

You might PM alfadriver if he doesn't find this thread.

AFAIK soldering O2 sensor wiring only applies between the sensor and the connector. I'd check it has a good ground, power, etc. ohm out the signal wires. Also apply dielectric grease to all the relevant plugs.

OK, so I did a run around the block doing a data stream on the scan tool and watching the o2 sensor voltages. B1S2 read 1v for about the first mile (after idling for a few min in the garage at 1v), then dropped to the ~.4-.8v range for the remainder of the drive. Which is about normal, from what I understand, correct?

So, I'm assuming, the reading dropped when the sensor came up to operating temp. I presume this means the heater circuit is working? Or should it have come up to temp more quickly?

Cleared the code when I got home, gonna see what happens. Any thoughts?

1v is really high for a cold O2. They read low until warmed up.

It would make sense, however, if the heater circuit had a bad ground and the heater was finding ground through the sensor. Which was a common failure on Chrysler O2s... the heater would find ground through the sensor circuit and the sensor reading on the scantool would be pegged rich. Read it with a voltmeter and it would be in the 5-12v range.

Chrysler is especially fun because they bias the O2 ground by 2.5v as an error checking strategy. So the O2 sensor reads 2.5-3.5v instead of 0-1v, on a voltmeter. But not all of them, only some of them, so you have to know what you're working with...