Oxygen Sensors - What NOT to do...

Well, I have never EVER recommended that anyone buy anything but genuine AC Delco O2 Sensors when it comes to GM vehicles. Then I go cheap because I really didn't have the money to buy the AC Delco parts. $60 later (for 2 upstream sensors) and my Bravada was sittin pretty. I jumped 3 mpg and life was good.

That was until today. Driving to a fire training exercize and it starts to miss and stumble. Wide Open Throttle (WOT) is fine, but just cruising, it's missing quite a bit. Just before I pull into the fire hall parking lot, the CEL (check engine light) pops up. DAMN IT ALL!!! Anyway, I run the codes and get 2 B1S1 codes for the O2 sensor. Damn things were brand new Borg Warner not even 2 months ago (less than 400 miles on them) and one is bad already. Get this, only had a 30 day warranty... Now I would have spent double to get the AC Delco parts (thanks to a local AC Delco retailer - from a dealer it would have been well over $200 for 2), but I would still be further ahead by now...

Anyway, for those of you that want to know, the AC Delco pn is AFS-105.

Anyway, just had to vent. I cleared the DTCs and will wait to see if they pop back up or not. On the test drive after the clear, all seemed well. I knew it was a sensor problem because after the CEL came on, it ran fine (ie. no missing or anything). That's because it goes to a predetermined fuel map when it has a sensor failure in the fuel delivery system (O2 Sensors, MAP sensor, MAF sensor, TPS sensor and Coolant Temp sensor).

I just got my blazer and that was one of the first things I have changed. I used the bosh oem which are the same ones the factory puts on. Make sure to put lots of ati-sieze.

Sorry ivannj, but Bosche is the ABSOLUTE WORST O2 sensor for GM vehicles. I stay away from them like they've got the plague. What GM puts on is AC Delco nothing else. The OEM part of the Bosche sensor is the fact that it can plug directly into the wiring harness, unlike their universal O2 sensors which require you to cut the connector off of your old O2 and splice it onto the universal sensor.

Also, lots of antiseize is bad as well. You just need enough to coat the threads on the sensor. Too much and you could end up with antiseize getting into the sensor and ruining it.

Also if no anti seize was used on the one you are removing don't get ahead of yourself and jerk on the wrench. You could just break the whole sensor bunge right out of the exhaust pipe as I did to mine. ooopoppps

IF you don't have anti-seize and even if you do. I always use a torch to heat the threads up. Helps in removing them a bunch.

And your sure your EGR isn't plugging up? It will also trip the O2 sensor codes.

EGR was just off and cleaned shortly after purchasing the truck (about 1500 miles ago). It was clean as a whistle inside (well, clean as a black soot lined whistle... no chunks or excessive buildup). The EGR valve wouldn't cause a lazy response from the sensor. I can graph the O2 sensor outputs (B1S1 vs B2S1). Comparing side to side, one side is cycling around twice as fast as the other. A faulty EGR valve should impact both of them the same. Plus, I have checked the EGR valve desired position as commanded by the PCM and the actual position as the sensor in the valve states and they are spot on all the time. If it was hanging open, the actual position would be different.

As with everytime that I replace O2 sensors, I heat the bung (as stated above) with a torch. Won't have to do that this time when I replace them as they were antiseized well and just off less than 2 months ago.

I would have gladly heated mine too but I was at a backwoods part store in the middle of nowhere at night.

Sounds like you got it covered. Hope those new O2 sensors fix your problem.


I know all about making repairs in the middle of now-where, coldest day of the year, or worst snow storm in years.

I can't believe you think bosh is a bad brand but you also think Royal purple is bad too lol. Just busting balls.

bosch will do the job just as well as any other but it's the lifespan where they fall short to delco and i wouldn't use wells as i have had a few of them were the screen cap thing that is actually on the inside of your pipe has fallen off. It's a funny thing when you take your o2 out to find somethings missing.

I can tell you one thing. I put in the AC Delco AFS105 sensors last night and logged a run. Then I compared it with a log from when I put in the 'new' Borg Warner O2s. HOLY FREAKIN' HECK Batman! What a difference in the response of the sensor. Night and day. The AC Delco actually has points between the peaks and valleys on the graph while the Borg Warner just hits the highs and lows. That is a great thing for the PCM to see.

Now, The B1S1 (driver's side upstream) was heat discolored (kind of a coppery color) up on the nut and was soot covered at the sensor tip. And B2S1 (passenger side upstream) was just like the day that I put it in... I didn't get a chance to take a look at the plugs, but it idles nice and smooth now where before it would idle really rough. Could the sensor malfunctioning cause bank 1 to run rich/hot (they don't seem to go together in my mind... normally rich = cooler, lean = hotter)... Or could the heater of just malfunctioned in the sensor and basically fried itself? I got a P0131 - B1S1 low voltage and P0134 - B1S1 Insufficient Activity... The low voltage means the sensor was running abnormally lean, and dumping fuel in like there's no tomorrow to try to compensate. I'll have to log some more runs later on this week to get a full picture of if I have a problem somewhere else in the system...

Have you checked out your temp sensor yet?

dont forget.... you could have a problem higher up than the O2 sensors, and this is causing the O2 sensor to crap out... you are probably running too rich, due to something else...


look into that first before you spend more on the O2 sensors..

good luck

Ok... Well damn it... I just had a nice long post and now it's gone... Somehow I got logged off and it disappeared.

Anyway, I am not ruling out a problem higher up, but the fact that I didn't have a problem before... I replaced the sensors as preventative maintenance. I scanned the computer and through some simple tests, found that both B1S1 and B2S1 were lazy. These 2 sensors combine with other sensors that tell the computer how much fuel to inject.

I have inspected the fuel system, running through all of the checks as stated on AllDataDIY.com, including leak down tests from full pressure (60-66psi) and 10psi. Both checked out perfectly fine. I also strapped my pressure gauge to the hood and test drove it with it on there. Everything was well within specs there as well. If I had a fuel system problem, it would not have passed all of those tests. With the old O2 sensors, the LTFT were out of whack on B1. With the new sensors, they are both holding steady right with each other. Also, the average injector pulse widths (ms) were reading high on B1 to try to compensate for the abnormally lean condition that the sensor was reading, while the B2 IPW was fine. Once I replaced the sensors, both IPWs were almost identical (within afew tenths of a millisecond).

I would think that if it was something like a temp sensor (or any other sensor that effects fuel delivery) that it would effect both sides of the engine the same. The B2S1 sensor that I pulled last night was working ok, still not as active as the Delco's, but it was working.

Ok... I'm going to post this now before it disappears again... I'll add anything else later

Oh... Another thing. Are there separate temp sensors for the gauge and what the computer needs to see or are they one in the same? My gauge has been spot on and always checks out with what is displayed on my computer when I tap in.

*EDIT - I do have 2 temp sensors. One for the gauge and one for the PCM. I haven't check that recently... Probably haven't in about a month. I'll check it tonight as well as whip out my infrared thermometer and grab a rough estimate of temperature from the t-stat area.

There are usually two separate temp sensors. On for the gauge on the thermostat and one for the ecm on the head.( mine is left bank).
Now my ? is this "My 1st gen only haas one o2 just before the cat- would it be possible, efficient, and worth my effort to weld close that bunge and put noe just past the exhaust manifold on each side. Most V8's have one in that location but just on one side. Would the comp read accuratly with the two in place and wired together. (Is that all making sense.)
Delete the original location and move to each end of the Y-pipe with joined wiring. If I do this I will likely use the heated ones as well.
I know that most 2nd gens use three but they are not all tied in together. I just want two but are tied together.
OR would I be better off just slapping a heated unit in the original location and just leave it at that.

All this is because I will be changing mine in the very near future (about that time again) and I just want to set it up as best as possible all in one clean swipe.




oh and I"M NOT TRYNG TO STEAL YOUR THREAD OR ANYTHING IT IS JUST THAT YOU APPEAR TO BE PRETTY GOOD WITH THE ELECTRICALS SO I THOUGHT I WOULD ASK.

swartllk,
You have some pretty cool tools.

I scanned everything again today and logged a run.

ECT was spot on 189-199 *F which is in the right frame for a 195*F stat.

So far everything points towards a crapped out sensor.
If anyone wants to see what data I pulled, and knows how to decypher it, have at it.
11.23.05_brav.02.xls

I transferred the data into Excel. I can also replay the data on my laptop so I really just drive when logging and review it after the fact.

I can get a LOT of different data. I have the GM Enhanced data as well. For a complete listing of what I can do, go to Autotap.com (GM Enhanced Data listing).

chevylane2003, I wouldn't go moving it. Due to the fact that you aren't really sequential injection, it uses a batch fire method and the banks are not independant. That is the reason why there is only one. You really wouldn't benefit from having 2 O2 sensors with the current injection and control system.

Didn't really figure it would be useful. Was just asking so I can prep for the work. I guess that I will just get a heated one and leave it at that. thanks

Stepping up to a HO2 would definitely cut the open loop time down. You'd have to run an extra wire and I'm not sure if it's 5v or 12v, but I guess it depends on the sensor.

Running the hot and cold wires are no biggy. All I have to do is figure out how hot it gets to be (5 or 12 volts). Maybe someone here has a part number for a 12v one. I am running low on 5v outputs and really don't want to relay another one. I already need to find a central nervous system to put all my existing relays and fuses as it is. I have like 10 fused relays running around under my hood already. Being a Blazer I could relocate them all to the tool compartment were the spare tire would be but that is a lot of wire to run that far and back. I would loose a bit of wattage doing that. I could run a heavy cable back there to a post and branch off it for all the hot wires to the relays but there is still the loses to get all that back to the acc. Where could I even run that much wire. There is too much to run them all through the interior so it would have to be under the whole truck. If I where to put it all there I would have to heavely insulate the cover to keep from hearing all the clicks and clacks from the relays. It would sound like an electric drum set back there. It could all go under the dash but it is already pretty full from all the interior relay work. To top it all off I need the glove box and console to remain as they are.

When, mileage wise, is a good time to change the O2 sensors? I've got 26500 on my 03, and i would like to change the sensor, but i'm wondering if it would be a waste of money, thanks. Also, i was looking under my truck, and i noticed that there is a dent in the exhaust pipe right near the sensor. I don't know if this is a factory installed dent, or a previous owner installed dent. Thanks.

The only proper way to diagnose a bad O2 sensor is with a scan tool (or oscilliscope) and watching the response of the sensor. How is your fuel mileage? Have you checked you plugs lately? At 26xxx miles, I wouldn't be worried at all. My Bravada made it to 75k miles on the originals (atleast to my knowledge they are the originals). As a check at 26k miles, I would just pull the plugs and check the gaps.

I see you are running a K&N filter, you should pull the MAF sensor (between the air filter and the throttle body) and clean it thoroughly. Be very careful with it, but with any aftermarket oiled reuseable filter, the oils coat the MAF sensor causing improper readings which could cause MANY other problems.

Once again, your knowledge astounds me. Thats a good call on the K&N filter thing. Especially since it would let extra dirt in, it will just cling to the oil. How would you recommend cleaning the sensor? And how often, like every time i change the oil? Thanks alot, you're the best.