Well, I learned my lesson today with Bosch O2 sensors. Bought a new one about 3 months ago (part# 13942) and it's already dead. It's a heated O2. Does anyone here know of a good substitute? It needs to be heated and definitely not a Bosch sensor.

A/Cs

...as in Delco?

try Carquest


Standard Automotive products 02 sensor

from a 1993 chevy/gmc TBI 350 truck 3 wire hookup

Carquest has gone to bosch as their supplier for O2 sensor's. This is straight from the Carquest Tech Rep that was at our shop about 6+ months ago.... Also, in their listing in their catalog, they list O2 sensors as "Oxygen Sensors, Carquest/Bosch" Therefore, if you don't like the bosch, don't go to carquest!

HTH

For the best 3-wire heated O2 sensor get the AC Delco AFS-74. This is a GM sensor that is tuned to the GM ECM PID fueling algorithm. May cost a little more, but well worth it.

Here are some additional GM part numbers for that sensor:

25165312

25312179

RBob.

Rrob, Does that sensor still have 450mv stoich.

The second part number registers at gmpartsdirect.com for $35, which according to them is almost 50% off retail. Knowing them though, shipping will be as much as the sensor

I just got done w/ my new motor and I put one of those bosh o2 sesors in.
While getting data from "TunerProRT" I notice the o2 has a lot of 000mV. Does this mean its crap or does the ecm do this sort of thing..??

AFIK, the O2 voltages should fluctuate between ~0.2V - ~0.8V. This is called cross counting. The faster is cross counts, the better the O2 sensor (from what I've read about this). My Bosch stopped cross counting and would hang at around 0.9V. Over the weekend I bought a Borg Warner O2 sensor because none of the local parts vendors carried the Delco sensors. I installed the BW sensor and it doesn't cross count at all. So now I'm thinking I bought a bad sensor. At least my Bosch sensor CC'd when it was new.

BTW, anyone here know how the ECM uses the CC to determine rich or lean?

AFAIK, B/W name is owned by Echlin now and used to market Echlin products.

Do you have a p/n for the connector?

http://sethirdgen.org/HO2S.htm

12126012 is the part number i got for the weatherpack connector

Thanks!

I was already using a 3 wire O2, but it had a different connector than the 25312179.

RBob, can you expand a little on your statement?

Thanks

Bill

Not all O2 sensors are created equal. The biggest difference I have found with this particular sensor (AFS-74) is that the AFR vs output voltage is not as steep as other sensors. This allows the fueling algorithm to have better control over the actual AFR.

Having driven vehicles and data logging the O2 sensor values of various sensors such as Ford (Bosch) , Chrysler (NGK), and a universal Bosch, the AFS-74 gave the best drivability and most consistant output voltage curves.

RBob.

Does anyone know the answer to this question...??
Is this normal for the o2 reading to stay @ 0volts..??

No, not normal. When in lean cruise mode will go to the double digits (< 100mV), but not 0.

RBob.

So, have you ever heard of this..?? No the wire is not broken. At least that I can see. Could this be the O2 sensor itself....who knows..!!!

Also one time I noticed the TPS sensor went to max voltage and I was not on the gas. I may have ecu issue.

THank you for the input. I will def look into it more. Right now I took apart the tilt to tighten and found it all broke so its gonna be down fer a week or so.

In the past I priced a pig tail from GM. . it was more then I paid for the sensor. So, I just popped the pins out of the sensor connector and soldered to them.

Recently I was given a housing and some pins, the housing is a Help Rack part #85319 (Motor Mite). The pins are loose so no number. Can also pop the pins & wires from another connector and insert them into this housing.

For those recycle yard scavengers the AFS-74 sensor was used in a whole bunch of '92 through '94 GM cars & trucks. The '90-'95 Corvette also used it. May be able to find both a pig tail and a sensor (C/K/R/V/P pickups, astro vans, suburban, G van, envoy, jimmy, sonoma, safari, yukan, bravada, cutlass, vista cruiser, F85, grand prix).


RBob.

I hate to even get into this, but the first thing you should check in this case is that the definition file is correct for your application. Sounds like you could be tieing the 02 to the incorrect byte in the aldl datastream. That would be the *first* thing to check.

I will check this weekend.....
I am pretty sure this is correct thou.....
Everything else works fine.... but, you are the man... And I will check.
Also........ Is thier a way to tap a potentiometer in line w/ o2 to get the car to force car to run lean..?? I didn't pass emissions . I blew a 273 hydrocarbons and limit is 220. I need to get the flash prom ordered and do it the right way..

As far as the pot inline with O2 wire, I don't think it will work (not with out more circitry) You want the computer to think the exhaust is richer than it is to force a lean condition. IE the mv reading would have to be higher than actual to force a lean condition. You would have to increase the mv somehow, not decrease it with a pot. Now a simple 2:1 coil and a pot........???

I think tunning is the proper way though.

High HC is usually a lean miss, or ignition problem.
Thou, old oil, and bad cat can also be causes.

A pot on the O2 line ain't going to work. You can change the voltage thresholds in the code to move the AFR a little or work things out with a several Op-Amp comparators, and digitial switches, but it's alot of work. I had built an O2 simulator for the ecm bench to give the right feedback to the ecm, but it's alot of work.

I haven't checked the part numbers, but I can tell you for sure that the O2 sensor in my '92 K1500 is the same one that is in my '87 and '83 'birds (I compared the '87 and '92 side by side). I'd be very surprised if any of them were replaced at any point

Hmmm, that just may be, in checking that particular vehicle the Bosch cat lists both a 1 wire and a 3 wire as possible sensors. This is for the 4.3l, 5.0l, and 5.7l. The 7.4l is listed as a 1 wire sensor.

I wonder if it is a california thing for the 3 wire sensors? With the other states getting a 1 wire?

My previous post was based on the sensor application chart. IOW, the list of vehicles that use that sensor.

OK, just checked a '93 Grand Prix, yep shows the 3.1l calif model gets a 3 wire, while the others get a 1 wire. Didn't specify this for the K1500 entry.

Just when I think I'm on to something. . .

RBob.

I wouldn't doubt the lean miss. I had to adv the timing to keep it from knocking when load goes up. It drives fine as long as I granny drive it. I am going to order the "prominator" as soon as I get some loot. I was just hoping for a cheezy way to pass emissions. As far as old oil and cat goes. I've got but 1200 miles on this new motor and I've changed oil in her 3 times already.... 100 mile breakin and 499 )pre-kick her in the rump to see what shes got) mile.

The main concern I had was the O2 voltage going to zero on the RT software. But, i've still yet to pull the ecm to make sure Im using the right data file (as described above).

Guys, sorry to change the subject slightly, but I was wondering if anyone has ever used KEM brand O2 sensors? I can get them reasonably priced locally and was wondering if it's worth using them or if I should just order some AC Delco sensors?

Thanks,
Rob

OK folks, I've been told in the exhaust forum that my installation of a heated o2 sensor due to adding headers into my 91 Firebird is all wrong. Since this is really ECM related, and how it handles the sensor output, I thought I'd check in with the experts over here...

Here's what I did:

1. I cut the wiring harness-to-sensor connector off the old sensor
2. Spliced the original sensor connector onto a new connector that mates with the new 3 wire sensor connector using a short piece of the old sensor wire (so that I could still plug in a one-wire if I wanted/needed to, or be able to easily change the 3 wire sensor if I needed to)
3. Connected the other two wires on the new sensor via the extra mated connector to keyed power and ground (again, so I could easily change 3 wire o2 sensors.

The gents in the exhaust forum are saying the original wire on the original sensor was "special", something about a fixed resistance, and had to remain unchanged for the sensor to be read correctly by the ECM. Furthermore, besides getting bad readings, they state this would probably also fry the new sensor.

I've been watching stuff via Datamaster, and I'm still getting CC's on the o2 sensor, so I don't see how it's fried. But in your opinion or experience is it valid that I'm no longer getting "good" o2 sensor outputs?

I'm still at a loss on how this could happen.... but look forward to being educated.

One last question - I've notice RBob doesn't like the Bosch sensors (I used theirs), and prefers the Delco AFS-74. If it makes a difference, I definitely switch it over (once I verify my wiring above). I've seen 2 part numbers for this sensor mentioned in this post - 25165312 and 25312179. Are both of these part numbers for the same sensor? The first number is not found on gmpartsdirect.com, the 2nd one is though.....

Sure hope you experts can help me learn some more on this topic. I plan to get into the DIY stuff myself before too much longer, but will try to read the primer on it first.

Have a Great Day, Y'all!


- Vern

you should be fine. the wire is not special, other than the fact that it is high-temp (wire and insulation) to prevent being melted. The impedance is so high on the oxygen sensor input that a few ohms of resistance in the line will do nothing to the signal. If that wire were special, it would be directly connected to the ecm somehow, there is more resistance(not repeatable either) within the connector pins.

Some other discussions spoke about the wideband sensor and how it obtains reference air through the insulation of the wires, which I still feel is not correct. I believe the reference air is obtained though the sleeving around the wires, not within the wires, but who knows for sure, the datasheets are not clear. Seems the wire would quickly corrode with ambient oxygen present, making the already small voltages and currents hard to measure accurately.

Hmm...you know, now that I think about it, there may be some problems, if the connection is made improperly. I said above the wire was High temp, and I think its a stainless steel wire, much like a coil resistance wire or fusible link. So the wire would have more resistance than copper. However, I don't think it has to do with the resistance as much as it has to due with dissimilar metals being connected together. The factory crimp may have unique plating to prevent electrolysis(galvanic reaction) on that junction, like nickel plating or something. when two dissimilar metals are touched together, they generate a weak voltage(and current), so the crimp you made could have been copper or tin or nickel or silver plated(or even other metals) and it might be that if you used one that would generate very little voltage, the sensor operation would still be OK, but one with the wrong plating could make the sensor completely useless. I'll have to split one of the sensors apart when I get home and take a close look at the sensor crimp. I'll also have to revisit my chem books to see what kind of voltage are generated by different metals......HMMM....interesting.....

I beleive this could cause problems.. There will be a voltage created if disimilar metels are used.. It could be small, or large depending on the metals. More than likely small, as to get a large voltage (more than like 0.2) you would have to have a highly corrosive metal, like aluminuim, or magnesium.

I hope someone with experience on this matter jumps in.....this is interesting.

I did find some information regarding splicing O2 wires. On Bosch's website, they have some detailed instructions for splicing a universal sensor into the existing wire harness. They use Posi-lock style connections, that have an outer weatherproof casing. So just the fact that they tell you to cut the wires means that it is possible. Now, the posi-locks they have could have the same metal as the wire on its contacts, making the connection non-galvanic.

http://www.boschusa.com/images/OxygenSensorInstall.pdf

another thing I found is a cool chart which show how differnt metals react. no specific numbers but it shows the potential as a list of metals.....Also, moisture is also needed to generate any voltages usually, or so it says....

http://www.metal-mart.com/Guides/Galvanic.htm

The dissimilar metals thing is the problem on more then one level. To start with the stainless that they use refuses to be soldered well, and you may get some galvanic action between dissimilar metals.

FWIW, the truck O2 sensor appears to be a '92 vs '93 and up

When wiring in a sensor w/o a connector I like to solder to the pin. IOW, pop the pin(s) out of the connector, if needed cut the pin a tad shorter, then solder the new wires to the pin.

I've had some sensors that I could not solder to the wires with lead/tin solder (ended up using a solder with some silver in it), then others that soldered easily.

Now I just leave the pins on and solder to them.

RBob.

Might try looking for GM TSBs when working with Delco sensors. According to them the standard O2 sensors need to breath thur the wire bundle. Mechanical connection for the wires, and no heatshrink sealing. This goes back to about 1986.

I've had three bad Bosch O2 sensors in a row, and on the DIY-EFI board there was alot of discussion about a known internal grounding problem in their sensors, several years ago.

I should be fine on the "breathable" wire covering. I did not mess with the pigtail on the 3-wiew O2 sensor. What I did was create a little conversion mini-harness with the connector that matched the 3-wire sensor pigtail on one end and a connector that matches the one-wire harnes on the other end for the signal feed and took the other 2 wires in my conversion harness to power and ground. That way I can change O2 sensors easily (not too sure I was clear on this explanation earlier) by just unplugging it and plugging in another 3 wire sensor. If I decide to try a one wire I just unplug my conversion harness and plug in the one wire sensor.

So the only difference signal-wise is a short piece (6" or so) of extra wire from an original one wire O2 sensor that is actually carrying the O2 signal.

And bash me all you want, but I did use a crimp on solderless connector for the signal wire in my conversion mini-harness.

So, provided the original one-wire sensor was not some kind of special wire (sure looked like standard plastic coated stuff) I only added 6" of wire's worth of resistance to the circuit. I haven't checked the length of the pigtail on a 3 wire vs. a one wire, but I doubt they are identical in length either, which would imply that I should be fine.

Does all this sound reasonable, or am I missing somethng here? (Anything is possible!!!!!)

Have a Great Day!!!


- Vern

Follow up Q to the AFS-74 sensor AC Delco version. 2 GM p/n's I am looking at are:

25312179 list $76.44 GMparts Direct 38.22

25176708 list $128.96 GMparts Direct 64.47

So which one is the AFS-74?? What are the differences between the 2? Cost is a big difference. Maybe one is a 4-wire?

That would be the 25312179.

Use AC Delco's Competitive Cross Reference to find ACD part numbers for many/most GM parts.

See image. . .

RBob.

Thanks guys, just the info that i needed.

Anyone have any problems from buying O2 sensors from that online parts store, GMParts Direct? The price is half the cost of what the dealer gets!

Mine came in a couple of days.....

Keep in mind they charge a flat 20% for shipping. Still should come out a little cheaper than what the dealer wants though.

hey did gm or ac delco put any part num,bers directly onto the sensor anywhere ??

stamped numers or anything ??

thanks

Notice the AC and the Delco logo's in the sticker
http://www.yankeelady.com/dobbins/gmallgm.gif

I've order a AC Delco heated 3-wire O2 sensor from www.gmpartsdirect.com and it was made by bosch.

I did some digging and came across this. The AC Delco cross ref. is a great link. I need to use a heated O2 sensor in a downpipe and have used the AFS-74 in the past. The AFS-75 is listed for a lot more vehicles.
The AFS-74 seems like it was installed in more batch EFI systems and the AFS-75 was put in more sequention EFI systems.
Anyone have an idea of what is different between them? Response time? Non-linearity differences?

EDIT: What makes wonder is the O2 negative sense line that can be tied directly to the ECM "O2 ground".

Man, those mating connectors are a rip off. Going to check a generic 3 terminal weatherpack without the pigtail to see how it fits. Looks the same but maybe keyed different and need a little drilling action.

I've been hearing of that happening.....

I guess it's time you step up to a WB.
While intially $$, in the long term, they'll save you a fair amount of $$ in gas.