Date:         Sat, 9 Oct 1999 10:16:56 +0200
Reply-To:     Darie Duclos <Darie.Duclos@UNIFR.CH>
Sender:       Vanagon Mailing List <vanagon@gerry.vanagon.com>
From:         Darie Duclos <Darie.Duclos@UNIFR.CH>
Subject:      Re: 1/4 mile syndrome is back!
Content-Type: text/plain; charset=us-ascii

Hi again,

It's the oxygen sensor! I waited all week for the problem to
show up again, and when it did last night, I disconnected the
O2 sensor. Without the O2 sensor, the problem went away! In fact,
the bus didn't even run particularly bad. I can't say for sure
that I could see any difference with the way it usually runs.
Connected it again, the problem was right back. So I drove the
rest of the 120km trip with it disconnected. Didn't even seem to
use more fuel than usual (15l/100km).

Now, as I understand it, especially since the problem is
intermittant, it could be the wiring in and around the O2 sensor
and not the sensor itself. The oxygen sensor is kind of expensive
to replace, especially for no reason. How can I test the sensor
itself to incriminate it or the wiring?

I received some pages about electrical testing in the Bentley
manual from William Dummitt (thanks again William), but this
says only how to test the wiring between the O2 sensor and the
ECU (right?) My friend's logic was that this wiring should not
be at fault: if the wire was broken along the way and sometimes
caused the O2 sensor to be ignored, then the effect should be
the same as when the sensor is disconnected altogether, right?
Suppose, on the contrary, that my O2 sensor is defective but the
wire is broken and the O2 sensor data is ignored most of the time.
Only when everything gets hot and the metal expands, the wire makes
contact and the faulty O2 sensor sends bad information to the
computer. Is this too far fetched?

Otherwise a few of you mentioned checking out the grounding
wires. Only I can't find the ones you mention: I didn't see
anything green, nor a coaxial cable (a mechanic even said when
I mentioned the coaxial that I didn't have one of those..)
I'm going to look at it again today, with my Haynes manual
in hand.

Of course many people mentioned the Digitool, and I'll have
to see how complicated it is to get one of those to Switzerland,
but in this case, would this tool be able to tell me anything
more that what I now know, i.e. the O2 sensor data is
intermittantly wrong?

Thank you for all the useful diagnostics suggestions. The most
frequent and most likely were the O2 sensor and the AFM. Makes
sense: one of the two sends wrong information to the ECU, so
suddenly the mix is much too rich. The temperature II sensor
was also a candidate.

By the way, can someone confirm that there is no point is performing
any electrical tests unless the problem is happening at the time?
Would seem logical to me, but it makes things difficult because,
as Murphy would have it, the problem only occurs when I don't
want it to.

Thanks. I'll keep you posted on further developments. Hope I'm
not boring you to death with this.
Bye!

Darie
Multicolor '86 Westy (just passed Swiss bi-annual inspection! Yeah!)