The vehicle in question is my 02 Silverado 2500HD DD packmule with the 6.0 liter. It has been running bad at idle and off idle until it warms and hits closed loop. Once it makes it to closed loop it runs pretty good. The internet said it was a common problem, caused by intake manifold gaskets failing causing a vacuum leak. I quick scan shows bank 2 o2 sensor reading wonky and fuel trim way off. Looks like a vacuum leak to me, so I change the gaskets, but I don't see anything wrong with the old gaskets. I get it back together and no change.
So today I hook the scan tool back up and it has a code for o2 sensor bank 2 sensor 1 insufficient activity. I check live data on the o2 sensors with the key on engine off. The other 3 sensors all read 450 mv, b2s1 is reading 50 mv. So bad o2 sensor I figure. I couldn't remember which sensor b2s1was for sure but I thought it was the front passenger side. I figured to verify that I would just unhook that o2 sensor and see which one was reading 0 mv on the scan tool. This is where it gets weird. With the o2 unhooked it now reads 125 mv, and the truck runs better. Whats up with that? Is it normal for the pcm to have a reading with the sensor unhooked?
Have I stumped the horde?
Well, no experience from similar trucks or motors, but from a purely electrical standpoint it makes some sense.
If the bad o2 is effectively shorted (no resistance) then it will always show 0V across it. If you unplug it, you break the short, and now there is a possibility you can have some voltage across the two ends of the wire. .125 V seems like a value that you may be able to generate via interference with a long o2 wire that isn't grounded to the exhaust pipe anymore.
Long story short, if the truck runs better with the IDed bad o2 unplugged, I'd say you nailed your bad part.
I would agree with rcutclif on his points. Also, I know the GM motors have a base map they default to if an O2 sensor disconnects. I have seen them run without any O2 sensors attached at all.
In best practice you would replace all the O2 sensors at the same time with the same brand/model. Theoretically they would be calibrated the same.
Also, one area I would look at is if you have a exhaust leak at the headers. Exhaust leaks can cause some wonk O2 sensor readings.
My go to O2 diagnostic is to just unplug it and see if it runs better on the base map.
After a day of driving with it unplugged i can say that it runs better cold, but worse when warm. Does anybody think its not the o2 sensor?
Try unplugging both upstream O2s, getting data from only one might cause issues, probably best to replace them in pairs anyway. If that doesn't make it behave closed loop I'd suspect multiple failures or possibly a goofy ECU. Check all the other live data and make sure it makes sense, also check fuel pressure.
Also remember that your trim tables will be massively out of whack when you do replace the sensors, don't forget to reset the ECU when you do that.
My 2000 6.0 has displayed similar symptoms twice. Scanning with HPtuners shows that once the CEL comes on from the fault, it does go into a pre-set map. No more fluctuating o2 activity after the code is set. Both times a new o2 sensor fixed the problem. First time was bank 1 sensor 1, second time about a year later was bank 2, sensor 1.
unplug the other upstream see if it defaults to the same value? if not scrutinize the wiring on the sensor with the fault.
I suspected vacuum leak but after changing the intake gaskets i sprayed carb cleaner all over and got no change, and secondly bank one doesnt indicate a lean condition.
Just in case anybody suffering from the same problem were to stumble upon this thread I figured the polite thing to do would be post up the conclusion to this story. I installed a new o2 sensor last night and that seems to have fixed it. There was also an open in one of the wires but it was for the heater circuit of the sensor, so I don't think that was the cause, but it got fixed also last night. It took a few minutes for the ecm to adjust to having a working o2 sensor in there, but then both upstream sensors were happily bouncing the voltages up and down, averaging around 450mv, just as intended. The truck is running better than it has in a long time.
