Re: O2 sensors

[John Larson <jdlarson@ix.netcom.com>]

Good morning. Orin, your reply makes more sense than Brendon's, but the first
section of both seems to unilaterally require the abandonment of the scientific
method. I will grudgingly yield to many of your points, as they appear to be based
on more than just visual observation, but I have a couple of questions which remain
unanswered.
First, the differences in both wholesale and over the counter prices cannot be
explained by the differences in connectors and wire length, even accounting for a
small change in the assembly line to allow for limited production runs of a
particular type. I can see 20 or 30 bucks, but not well over a hundred.
Second, if all were the same, Bosch would not bother to manufacture 2 different (and
visually identical) single wire universal sensors.
Third, Bosch would produce and market a relatively low cost multiple wire sensor,
and market it to the "BAPS" system of parts specialists as they do all their
products. I'm pretty sure they'd rather sell a universal sensor at a lower price
than make no sale at all. I have yet to see any information to that effect.
Fourth, federal lay may prohibit the installation of a non-approved unit in a
pollution controlled motor vehicle. In California, where I live, any pollution
control related part installed on a particular vehicle must be approved by the Air
Resources Board for that vehicle. We're often made fun of by folks from other
states, but our air IS getting cleaner, and fuel economy HAS gone up, not just for
us, but for all of you. Even with several hundred fold the number of motor vehicles
in LA compared to the late 1950s, the ratio of relatively clear days to really bad
ones has improved dramatically.
These considerations effectively prevent me from concluding that it is "established
fact" that these units are all interchangeable. It is my obligation to provide my
customers with the correct resolution to their automotive problems, and "cheaper" is
not the first, or even the second factor on the list of criteria. The terms "best",
"cost effective", and "correct" all come to mind. I'd much sooner trust the R&D
departments at VAG, Bosch, and the other major manufacturers than some snake oil
salesman or the semi cretin at the parts counter of the all night mass marketer
parts store. In my experience, "cheaper" is often far from cheaper, in the long run.

To answer your point about longevity, the mileage figures mentioned in my post
reflect the specifications by part number, not application, so no matter what
vehicle the unit fits, the recommended replacement interval remains the same,
truck/car, turbo/non-turbo.
With regard to the 5 wire unit, of which I was not aware, you might be interested to
know that Toyota and Jeep use a sensor in some applications that operates on a
resistance principle rather than a voltage signal.
Thank you for your time and interest, and I have learned from your well written
input.
John
Orin Eman wrote:

> > "Established fact"? By whom, and how? By what standards?
>
> Seem to recall from someone reading the numbers stamped on the
> side of the sensor...
>
> > Are you aware, that in addition to 1 and 3 wire units, there are 2 and 4 wire
>
> 2 wire is a 1 wire with the ground connection brought out separately
> rather than just bonded to the case.  Similarly for 3/4 wire.
>
> > sensors. Did you know that some sensors are rated for 30K miles, some for 50K,
> > 60K and 100K? Have you ever wondered why there are multiple numbers of 3 wire
>
> For the same car?  Turbo cars tend to eat them up faster.  Wouldn't be
> surprised if the absolute same sensor is speced for 30k miles in one
> car, 50k on another etc..
>
> > sensors specified for various years and models of Audis? Wouldn't it have been
> > cheaper for Audi to specify a single design/part number? Without further
>
> Different length of wire and/or connectors makes most of the differences.
>
> > investigation, I would wager that each number specified corresponds with a
> > different emissions system, i.e., CIS, CIS-E, Motronic, etc.
>
> Aging effects and temperature account for more variation.  The basic chemistry
> leads to the curve given in the Bosch books.  I've disassembled some of
> the ECU code.  It just looks a voltage in one of three ranges - continuously
> around .45 V (in which case the sensor is dead), above this range
> (rich) and below this range (lean).  That's all.
>
> > I'm convinced, having worked on Lambda systems since 1976, that it's wise to
> > stick to the products specified for each application. And that's what I am
> > going to do, both on my own cars, and those of my customers.
>
> My car is happy with the GM application Echlin I put in... had the classic
> symptoms of a dying O2 sensor - slow transitions, poor gas milage.
> Back to its normal gas milage afterwards.  It _may_ age faster than
> a Bosch, but new out of the box, it behaved just as expected.  Really,
> look at the curves vs temperature... without knowing EGT, the ECU doesn't stand
> a chance of using a regular 1,2,3 or 4 wire sensor for anything but
> rich/lean detection.
>
> John, I do see your point - maybe the generics age faster - and
> that uncertainty is enough for some... For me, give me the generic
> and when my dancing lights O2 sensor monitor gets slow and the
> gas milage goes down, then I'll worry.
>
> BTW, there are 5 wire sensors used in some Hondas too.  These are true
> wide range sensors when driven by the correct circuitry... do a patent
> search on wide range oxygen sensor or similar for some good bedtime
> reading.
>
> Orin.