OXYGEN SENSOR

Original Archive Name: oxysensor
Derived from ND BBS OXYSENS.TXT
Last update by Jan: April 27, 2000

Here are a collection of notes on the oxygen sensor, what it does, how to replace it, how to diagnose some related problems, how to turn of the red light that comes on after 30/60k miles.

Briefly, your oxy sensor (aka Lambda sensor) sniffs your exhausts for oxygen content, and feeds back this value in the form of a voltage to the car's ECU. The ECU in turn adjusts the fuel delivery to meet the desired fuel to air mixture value. This is also why this is called a closed loop or feedback system. Just like in the BMW commercials, VWs tune themselves, and have been for quite some time...

Oxysensors wear out, and if they do, it usually result in a rough idle, poor engine performance and higher fuel consumptions. Besides these annoying side effects they may also cause your catalytic converter to burn up or loosen up. A broken up catalytic converter sounds like a couple of coins rattling in a coffee can. I call it the death rattle. You do not want this to happen, because the cat conv., depending on the car, may cost up to US$1000 to replace, and in many states like CA it is illegal to buy a used one. Some just gut it with a crow bar (the ceramic honey comb is surprisingly strong), but that's very illegal (US$20 000 fine if caught) and will result in more pollution.
Newer cars actually expect the slight cat back pressure and therefore end up running *worse* with the cat gutted.

To remind users that the oxysensor may be on its final snifs, VW installed a little box that contains an odometer that trips a little switch at either 30 or 60k miles. The older cars with non-heated sensors have to be changed around 30k, the newer cars use a heated sensor at a different location thereby lengthening their useful life to 60-100k miles.

Many just reset that odometer to save money w/o replacing the sensor, and instead burn up the cat costing 10x more. There are ways to check whether the oxy sensor is still good (it depends on your driving, they may last much longer, see Bentley and below), but if you want to be sure, replace it when needed.

VW uses different styles of oxy-sensors, from the 1 wire (cheap), 3 wire (2 are used to heat the sensor at start) up to the 4 wires used in the Corrados. Naturally, each wire adds about 40US$ to the cost, and some claim you can use a 1 or 3 wire sensor instead of a 3 or 4 wire respectively if you do not care for cold running behavior. I would not advice it, but if you drive a beater and have no cash...

[According Matt Heffner: In the 4 wire sensors, 3 of the wires are for the sensor, the extra wire is the shield for the signal wire. This shield wire prevents electrical noise from being induced into the signal wire as it passes through the engine compartment. The sheild wire is grounded at the O2 sensor harness bracket (passenger side motor mount).

It's a good idea to use a bit of anti-seize on the oxy-sensor threads before installation, but don't let any of it touch the sensor because the paste contains lead. Similarly, keep products containing Si away from the sensor. It's a very sensitive piece of equipment.

Jan

Other resources:

o Lambda Probe test for 4 wire oxysensors.
 

Copyright Notice (c) -- 1990-2000
All Rights Reserved

The information contained here is collectively copyrighted by the authors. The right to reproduce this is hereby given, provided it is copied intact, with the copyright notice inclusive. However, the authors explicitly prohibit selling this document, any of its parts, or any document which contains parts of this document. 


From: msirota@ee.rochester.edu (Mark Sirota) Subject: Re: Oxygen Sensor-85 GLIDate:
20 May 92 13:44:52 GMT

In article <19949@oasys.dt.navy.mil> odell@oasys.dt.navy.mil (Bernard O'Dell) writes:
> I also have a service manual, but it doesn't show where the sensor is > and how to shut the red light off-there has to be a button or ?? to > do this.
>
> Where the hell is the button or whatever to shut the red light off-- > big secret to enhance the VW service people to perform this big task?

If there was an FAQ posting for this group, this would certainly be on it.

The "magic button", for A2-chassis cars ('85-'92), is just behind the coolant resevoir. Remove the resevoir by lifting it up, and set it down out of the way somewhere. (You don't need to remove any hoses or anything. Don't do it when the coolant is too hot unless you have nomex gloves.)

Follow the speedometer cable up from the gearbox. It will enter a black box which is just behind where the coolant resevoir was, mounted to the firewall. On that black box is a white button. Push the white button until you hear a satisfying "click". (You have to push it pretty far, use a nail set or a pen or small screwdriver or a punch or something.) Voila, your light is off, after only about 15 seconds of work. Don't forget to put the coolant resevoir back.

It comes on every 30K miles, not every 50K, BTW.

On A1 cars, and on most Audis, the "button" is at the top left corner of the instrument cluster, set way back from the front panel. It's a bitch, much unlike the A2 cars. You have to fashion a little hook out of a coat hanger or something, then wiggle it aroudn in there for a few hours until you give up. Then go have dinner or something, come back out to the garage, try it for a little while longer, and eventually you'll get it. From lynx.unm.edu!umn.edu!news-feed-2.peachnet.edu!emory!europa.eng.gtefsd.com!howland.reston.ans.net!newsserver.jvnc.net!erenj.com!erenj.com!bdboyle Wed Aug 18 10:58:28 PDT 1993 Article: 19007 of rec.autos.vw
Newsgroups: rec.autos.vw
Path: lynx.unm.edu!umn.edu!news-feed-2.peachnet.edu!emory!europa.eng.gtefsd.com!howland.reston.ans.net!newsserver.jvnc.net!erenj.com!erenj.com!bdboyle From: bdboyle@erenj.com (Bryan D. Boyle) Subject: Re: Oxygen Sensor Light
Message-ID: <1993Aug17.153025.29880@erenj.com> Lines: 31
Sender: bdboyle@maverick1.erenj.com (Bryan D. Boyle) Nntp-Posting-Host: maverick1.erenj.com
Reply-To: bdboyle@erenj.com (Bryan D. Boyle) Organization: Exxon Research & Engineering References: <1993Aug17.082955.15656@walter.cray.com> Date: Tue, 17 Aug 1993 15:30:25 GMT

This is in the FAQ, however...

  1. remove dash fascia.
  2. take a long piece of STIFF wire about the diameter of a paper clip, bend the end 90 degrees, leaving a short leg of about 1/8": --|

  3.  

     
     
     

    | |

  4. look along the top of the speedo/tach cluster on the left hand side. You will see what looks like this: -----------

  5.  

     
     
     

    | | | | || | || | | || | || | |--------------------------------------------- | (speedometer | (tach) | |

  6. Take the tool, with the short tang facing upwards, and insert alongside the tab in the RIGHT opening, turn 90 degrees and pull out. What you are trying to do is to catch the tab you see sticking up and pull it forward. If you are successful, you will hear a "thunk".
  7. light should be off.
--
Bryan D. Boyle              |Physical: ER&E, Annandale, NJ 08801
#include <disclaimer>       |Logical: Cogito sum, ergo sum, cogito.
908 730 3338                |Virtual: bdboyle@erenj.com 
                Sometimes the dragon wins...
Nov 16 12:56:51 PST 1993 Article: 22757 of rec.autos.vw
Newsgroups: rec.autos.vw
Subject: Re: 84 Jetta GL cold idle problem
From: mark@wdcwdc.sps.mot.com (Mark Shaw) Date: Thu, 11 Nov 1993 22:34:09 GMT
Sender: mark@wdc (Mark Shaw)

In article, (Dave Gervais) writes:

|> I recently bought an 84 Jetta GL in the color of lovely polar silver. |>
|> Anyhow, when I first start this car up, and it's cold outside and |> the engine is cold, it idles strangely, beginning at 1000 rpm |> and staying there for the most part, taking a sharp dip until |> the engine almost stalls, then back up to 1000 rpm... it cycles |> through like this.
|>
|> When the engine is even remotely warm, it does not do this at all. |>
|> Is this just some kind of problem with the "cold start" fuel |> injection settings or something or something much greater ? |> Any knowledge would be appreciated...

Sounds like an oxygen sensor and/or thermoswitch problem. Either the oxygen sensor is degraded to the point that its response is too slow until it gets to a much hotter temp, or the thermoswitch is allowing the control box to lock onto the oxygen sensor output too early (even a good oxygen sensor needs to warmup first).

One quick check would be to find the thermoswitch (usually in the water hose coming/going to heater) which is a cylindrical pipe with two spade terminals out the side. With the engine cold, pull off the two terminals and measure the resistance of the thermoswitch itself - should be a dead short when cold. Then check it again after the engine has warmed up and the problem goes away - should now be open circuit. If the switch does not work, that may be your only problem.

Next try taking the two connectors from the thermoswitch and short them together. This should keep the car in the open-loop mode (ignoring oxygen sensor) and see if the idle speed is more stable (it may still vary due to a flakey idle stabilizer). If the engine runs stable (but at a different speed than normal) then the oxygen sensor probably bad.

The real check of the oxygen sensor involves the procedure in the Bentley manual to check the duty cycle operation of the fuel proportioning valve, which is not too difficult to do, but requires some fooling around with the fuel pump relay and a voltmeter or dwell meter.

If the above quick checks do not seem to indicate that the thermoswitch or oxygen sensor are at fault, then check for a faulty idle stabilizer. It is a small box in the rain tray below the windshield and next to the ignition module. I would suggest using the Bentley manual to locate these two units. The stabilizer can be bypassed by disconnecting it and then reconnecting the two connectors from the ignition module back together. If the problem goes away, then flakey stabilizer.

Final check would be for some form of vacuum leak in the manifold, but usually these just cause the idle to wander around real slowly.

I would bet that it's the oxygen sensor, especially if its an '84 model and hasn't been changed since new.

Hope these help.

mark

BTW, I may be reversed on the thermoswitch being shorted when cold and open on hot, so I'll recheck my Bentley tonight to be sure.

Thu Feb 3 13:36:01 PST 1994 Article: 26471 of rec.autos.vw
Newsgroups: rec.autos.vw
From: mark@wdcwdc.sps.mot.com (Mark Shaw)
Subject: Re: [W] CIS-Lambda OXS switch strangities
Sender: mark@wdc (Mark Shaw)
Nntp-Posting-Host: margay.sps.mot.com
Organization: Motorola Western MCU Design Center, Chandler Arizona

In article, (Peter Tong) writes:

Hi, Peter,

Interesting problem.

|> Recently my 2.0 has been slightly misfiring when starting right from the |> cold. It wouldn't misfire under load but only at no load/idle and low rpm |> conditions. Disconnected the oxy sensor temporarily to see if it was a |> fuel cycling problem - no problem there - still slightly misfired.

Disconnecting the O2 can be ambiguous to the controller depending on the input circuitry. The Bentley manual gives two tests which minimize this by either connecting the O2 lead from the controller to ground or to a 1.5 volt battery with its negative end on ground. The grounded case simulates a lean mixture, and the 1.5 volt input simulates a rich mixture (actually 0.9 volts would do it, but who has a 0.9 volt battery laying around:-). Anyway, leaving the lead open may give either case.

|> The misfiring would disappear after the car would warm up. |>
|> So anyways, just for the heck of it I just disconnected the OXS temp switch. |> This means the car goes into closed loop right off the bat during a cold |> start. Suprisingly the car runs like a top! Engine completely smoothed out |> and misfiring disappears.

I always get confused here. I thought that leaving the thermo switch disconnected would leave you in open-loop. In which case the baseline adjustment of the screw on the flow meter would give you the right mixture; and the O2 sensor would be giving you too lean a mixture if it heats up too slowly. But, like I said, I keep forgeting which is open/closed.

The interesting thing would be to check the duty-cycle signal (if K-Jetronic) or the modulation current (KE-Jetronic) to see where the mixture actually goes to. This would also be based on your baseline mixture adjustment. Without this, it is sort of a guess as to what is going on. If the O2 sensor is OK it should start cold from the "rich" side and slowly move down in voltage as it heats up. This would suggest a lean mixture would be generated if the sensor were kicked in before it warmed up. If you are suffering from too rich a mixture at start this would help, but usually a too rich mixture at start is the desired case for smoothness and stability in a cold engine. So I am puzzled.

Just some random thoughts....

mark

Mon Apr 4 12:21:59 PDT 1994 Article: 29311 of rec.autos.vw
From: maslin@CVMFaculty.MSState.Edu (William Maslin)
Newsgroups: rec.autos.vw
Subject: [W] A1 oxygen sensor replacement experience
Followup-To: rec.autos.vw
Date: Wed, 30 Mar 1994 13:39:12 -0500
Organization: Mississippi State University Lines: 44

Several weekends ago, I replaced the single wire oxygen sensor in my 82 Convertible. I'd read some postings in the past about what a nightmare it could be (torches, drilling, etc.) so I was ready for a struggle. However, it wasn't too bad so I thought I'd pass on some of the details in the hope that they might somebody else.

First, the weekend before, I sprayed Liquid Wrench (LW) semi-blindly around the base of the oxsens while the engine was warm (but not hot enough to burn my arm or instantly evaporate the LW).

On the appointed weekend, and as per the instructions in Bentley, I first removed the black plastic and rubber air ducting from the mixture control unit and the throttle body. You need to remove the duct, etc. so that you have more room to work. Even so, there isn't much room to work in. (BTW, if you attempt this, be careful not to damage the air duct parts or the big "hose" clamps).

Next, I sprayed LW around the base of the oxsens again. I tried various sockets, wrenches, etc. but, because the recessed location of the oxsens, I couldn't get any to fit it. So, I bit the bullet and went to Auto Zone and bought a Lisle Oxygen Sensor socket. This is a long 7/8" socket (3/8" drive) with a slot running its entire length for the oxsens wire to pass through. It fit perfectly. The cost was $10.

I attached a 3/8" rachet handle to the socket, put the handle in the 5 o'clock position (when viewed from the front of the car) and commenced to push. I pushed my guts out but couldn't get it to budge. I was just about to go get something to beat on the handle with when I decided that there was just enough room to place the handle in the 11 ( or was it 12 or 1?) o'clock position. This put the handle in the "pull" configuration. I gave it a firm pull and, to my surprise, the oxsens came loose without any complaint. Installation was the reverse of removal, as they say.

So, the bottom line is that I attribute my success to two things: the Lisle socket and placing the rachet handle in the "pull," rather than "push," configuration. I guess the Liquid Wrench didn't hurt, either. Of course, your mileage may vary...

Bill
maslin@cvmfaculty.msstate.edu 


78 Rabbit...since new! 82 Convertible...since last July!


Mon Apr 4 12:44:07 PDT 1994 Article: 29368 of rec.autos.vw
Newsgroups: rec.autos.vw
From: jtabersk@harp.aix.calpoly.edu (Jeffrey S Taberski)
Subject: Re: 84 GTI Mileage
Organization: California Polytechnic State University, San Luis Obispo R

In article <2nb085$m30@search01.news.aol.com>, Tas3 <tas3@aol.com> wrote:

>You need to replace your oxygen sensor. I had the same problem on my >'84 GTI.
>
>Tom Skrovanek

There is a way to test the o2 sensor. I don't know about you, but I can't afford to replace every "suspected" part on my car when it doesn't run right. At any rate, here's the procedure (short form)

  1. plug the pos lead of a DIGITAL voltmeter to the o2 sensor lead (don't unplug the connector).
  2. neg lead to the neg battery post
  3. run engine to full op temp.
  4. observe voltages on meter the voltages should fluctuate above and below .5 volts. >.5 indicates rich <.5 indicates lean. If no fluctuation occurs (stays above or below .5volts) then there is something else wrong (ie: your engine is running too rich or too lean). If the voltage is zero, check the wiring, else you have a bad o2 sensor. If the voltage SLOWLY fluctuates above and below .5 volts, then you have what is called a "lazy" o2 sensor and should be replaced.
Silicone is the biggest killer of o2 sensors, so don't spray silicone spray around a running engine or have it sucked into the intake and contaminate the o2 sensor. Also, use RTV that is rated oxygen sensor safe.

About that bad gas mileage...try cleaning the throttle body and the fuel distributor sensor plate...it does wonders for driveability, believe me. It mostly helps for hesitation and stalling, but it couldn't hurt!

hope this helps

Jeff "Certified California Smog Technician" -- 


=           Jeffrey Taberski                =   '61 Morris Minor 2dr =
=    Agricultural Engineering Student       =   '63 MGB roadster     =
=          "British car nut!"               =   '65 MGB roadster     =

Wed Apr 6 13:02:07 PDT 1994 Article: 29477 of rec.autos.vw
Newsgroups: rec.autos.vw
From: mark@wdcwdc.sps.mot.com (Mark Shaw) Subject: Re: 84 GTI Mileage
Message-ID: <1994Mar31.174515.23617@newsgate.sps.mot.com> Sender: mark@wdc (Mark Shaw)
Nntp-Posting-Host: margay.sps.mot.com
Organization: Motorola Western MCU Design Center, Chandler Arizona
Date: Thu, 31 Mar 1994 17:45:15 GMT
Lines: 39

In article, (Jeffrey S Taberski) writes: |> In article, Tas3 <tas3@aol.com> wrote: |> >You need to replace your oxygen sensor. I had the same problem on my |> >'84 GTI.
|>
|> There is a way to test the o2 sensor. I don't know about you, but I |> can't afford to replace every "suspected" part on my car when it doesn't |> run right. At any rate, here's the procedure (short form)

|>      1) plug the pos lead of a DIGITAL voltmeter to the o2 sensor lead
|>         (don't unplug the connector).
|>      2) neg lead to the neg battery post
|>      3) run engine to full op temp.
|>      4) observe voltages on meter
|> the voltages should fluctuate above and below .5 volts. >.5 indicates rich |> <.5 indicates lean. If no fluctuation occurs (stays above or below |> .5volts) then there is something else wrong (ie: your engine is running |> too rich or too lean). If the voltage is zero, check the wiring, else you |> have a bad o2 sensor. If the voltage SLOWLY fluctuates above and below |> .5 volts, then you have what is called a "lazy" o2 sensor and should be |> replaced.

I must caution against using a DIGITAL voltmeter on either the OX sensor or the duty-cycle output from the control. They will not do any damage, but their inherent sampling rate may mask the actual voltage or its cycling nature. Analog meters are best for these measurements.

Also, the procedure in the Bentley manual is preferred, since it also includes checks of the controller and thermo switch, as well. If the controller or thermo switch are faulty the actual air-fuel mixture may be fixed at the wrong level and the OX sensor will therefore be stuck at one extreme or the other. Changing the sensor will not correct the problem in this case.

Normal operation shows a cycling between 0.1 to 0.9 volts on the OX sensor at about 1-2 times per second (after it is warmed up). The duty-cycle output will mirror this change. Many low cost DIGITAL voltmeters have sample rates in this same range and cannot properly follow these swings.

Mark


Wed Aug 4 11:15:51 PDT 1993
Newsgroups: rec.autos.vw
From: mark@wdcwdc.sps.mot.com (Mark Shaw)
Subject: Re: [W] A little diagnostic help, please
Sender: mark@wdc (Mark Shaw)
Organization: Motorola Western MCU Design Center, Chandler Arizona
Date: Tue, 3 Aug 1993 00:23:42 GMT
Lines: 56

In article (Jerome Lanig) writes:
|> I've got a problem and would like some input. |> ............................
|> Would the Oxy/Lambda sensor cause all this? I've only |> had it for about 4K miles (110K on engine). Would |> running extremely rich for a long time cause the carbon |> build up, poor milage, and herky jerkiness?

You mention that the sensor was unconnected. In typical OXS circuits this would look like a rich mixture and result in the fuel modulation valve running to the "lean stop" which would certainly give the results you describe.

If it runs poorly with the oxygen sensor connected, then the sensor is either defective or very sluggish in operation. Normally the duty cycle for the fuel modualtion valve should move between 45-55% about every second if the OXS is good and heated up. The test connector is near the passenger side strut tower as some other poster mentioned.

If you connect the OXS and the duty cycle stays stuck at 50%, then check the thermo switch in the hose next to the engine block above the clutch housing. It should be open when thte water temp is greater than 80-90F. If it is still closed the OXS control will stay in open-loop and then you are stuck with whatever the fuel distributor has been set too.

Someone mentioned that if you check the OXS with a high impedance voltmeter and the voltage should be 0.45 vdc for the stoichiometric ratio. This is true in theory, but the sensor will not stay on this point, nor is it desireable to stay right on stoichiometric. The three-way catalytic converter reduces NOX well if it is slightly rich with some excess CO to create N2 and CO2. But it also wants to be slightly lean with excess oxygen to oxidize CO and HC. The best compromise is to quickly cycle between a slightly rich (L=0.99) and a slightly lean (L=1.01) and use the catalytic converter as a sort of oxygen capacitor.

Good OXS will respond quickly and cycle about the 50% duty cycle (45-55%). Old or clogged ones will wander more and therefore spend a lot of time in either too rich or too lean regions. Since the 3-way catalytic converter still has some significant reduction behavior off-ideal, you may not see this in a CO measurement.

It is fairly simple to follow the procedure in the Bentley Manual to test the OXS system. Another good reference is the "Bosch Fuel Injection & Engine Management" by Charles Probst. Available from Classic Motorbooks or the SAE.

I agree with Peter Tong's experience and recommendations. Change the sensor for about $40. You should notice a big difference in cold warmup and run stability.

Mark

From lynx.unm.edu!news.cs.indiana.edu!usenet.ucs.indiana.edu!vixen.cso.uiuc.edu!howland.reston.ans.net!europa.eng.gtefsd.com!uunet!gumby!calvin!ursa!ptong12 Thu Feb 3 12:47:56 PST 1994 Article: 26435 of rec.autos.vw
Newsgroups: rec.autos.vw
Path: lynx.unm.edu!news.cs.indiana.edu!usenet.ucs.indiana.edu!vixen.cso.uiuc.edu!howland.reston.ans.net!europa.eng.gtefsd.com!uunet!gumby!calvin!ursa!ptong12 From: ptong12@ursa.calvin.edu (Peter Tong) Subject: [W] CIS-Lambda OXS switch strangities Message-ID: <CKIBM6.7EK@calvin.edu>
Sender: news@calvin.edu (Netnews Punk)
Nntp-Posting-Host: ursa
Organization: Calvin College
References: <2ij82n$f6l@sousa.ako.dec.com> <1994Jan31.174111.23428@galileo.cc.rochester.edu> Date: Mon, 31 Jan 1994 18:34:53 GMT
Lines: 29

Recently my 2.0 has been slightly misfiring when starting right from the cold. It wouldn't misfire under load but only at no load/idle and low rpm conditions. Disconnected the oxy sensor temporarily to see if it was a fuel cycling problem - no problem there - still slightly misfired.

The misfiring would disappear after the car would warm up.

So anyways, just for the heck of it I just disconnected the OXS temp switch. This means the car goes into closed loop right off the bat during a cold start. Suprisingly the car runs like a top! Engine completely smoothed out and misfiring disappears. My question is regarding the control pressure regulator. Does anyone know if the amount of pressure allowed through it varies over its lifetime. IE does the diaphram "stretch"? My car ran fine with the oxs switch connected during the summer but now the oxs switch and the control pressure regulator in combination seem to supply an overly rich mixture to the 2.0. I do know that the 2.0 has lower idle vacuum than my 1.8 did. At idle manifold vacuum is right at 16" hg. On my 1.8 it was at 19" hg with the exact same cam. The GTI had a rod/stroke ratio of 1.67 vs. my 2.0s 1.55, and I'm currently running slightly retarded timing with the adj. cam sprocket. Could these changes make that much of a difference idle vacuum wise? Seem strange to me. Compression on the motor is up at the 200 level warm.

At any rate the car runs much better and creamier now. Amazing how a small change can create a personality change in the car....

Peter Tong
'82 2.0 8V Cabby
.

From uunet!uunet!airs!andrew Fri Mar 6 10:56:08 PST 1992 Article: 4121 of rec.autos.vw
Relay-Version: Unigraphics Division (EDS) Bulletin Board 10/13/90 VAX/VMS V5.5; site mdcbbs.com Path: mdcbbs.com!uunet!uunet!airs!andrew Newsgroups: rec.autos.vw
Subject: Re: Tuning Up 1987 Jetta GLI 16V Message-ID: <3716@airs.com>
From: andrew@airs.com (Andrew Evans)
Date: 4 Mar 92 18:19:28 GMT
Sender: news@airs.com
References: <1992Mar2.174159.1571@linus.mitre.org> <2MAR199215531198@zeus.tamu.edu> <kr5nu1INNh58@jethro.Corp.Sun.COM> <3MAR199217110465@zeus.tamu.edu> Keywords: tunu-up
Lines: 41

jmm2948@zeus.tamu.edu (Jeffrey M. Mayzurk) writes about O2 sensors:

> Yes, the are expensive... the single-wire type is about $60 and the > dual-wire with heating element is about $90. Definitely not cheap, > but it IS a regular maintenance item and needs to be changed at > regular intervals, especially with tightening emissions requirements.

I got a single-wire type for my Audi Coupe for $35 Import Parts Connection. They're a mailorder place in CA that advertises in European Car. I think Wolfsport also has them for cheap. Don't know about the price of heated sensors, though.

> For the guy with the GTi 16V that hunts at idle, the way test the > oxygen sensor is to disconnect it from the computer. If you follow > the lead from the exhaust manifold, you will find a connector inline. > Disconnect this and see if the idle improves. Make sure the car is > warmed up, however, because the sensor is disabled until the coolant > reaches a certain temperature anyways.

This may show the problem if the O2 sensor is really at fault, but it could also indicate another FI-related problem as well. You can test your O2 sensor by connecting a high-impedance digital voltmeter that has good resolution within the 0-1 volt range. The O2 sensor should put out around 0.45v, but the reading will fluctuate because the FI computer is continuously leaning out and enrichening the mixture and using the O2 sensor to determine the optimum mix. Usually, a dead sensor will put out low or no voltage.

To do this test, connect the positive lead of the meter to the single wire coming from the sensor to the FI computer. Leave it connected to the computer input as you do this. Some sensors also have a two-wire connector, which powers the heating element, but don't bother with it. Connect the negative lead of the meter to the (-) battery terminal, or to a known good ground such as the cylinder head.

If the sensor is still fine after 30K miles, leave it in, but do this check if your idle starts wandering up and down. A bad sensor can cause your fuel mileage to get worse, but the improper fuel/air ratio could also spell the death of your cat converter, so don't let it go.

Hope this helps people diagnosing those pesky O2 sensors.
 



From: jan@lipari.usc.edu (Jan Vandenbrande)
Subject: Testing oxygen sensors (fwd)

Forwarded message:
> From ti@apple!cindy.Altos.COM Wed May 2 02:24:49 1990 > Date: Mon, 30 Apr 90 20:05:38 PDT
> From: ti@apple!cindy.Altos.COM (Ti Kan) > Message-Id: <9005010305.AA04443@cindy.Altos.COM> > To: info-vw@ames.arc.nasa.gov
> Subject: Testing oxygen sensors
>
>
> I have a good way to test the oxygen sensor to share with everyone. > Since a new sensor is not cheap, you'll probably want to know that it > is really dead before investing time and money on a new one. >
> This home-brew test requires only a DC voltmeter, and two AA > batteries. This assumes the Bosch K-Jetronic with Lambda or the > KE-Jetronic (virtually every fuel-injected, watercooled VW and Audi > from 1980-81) system.
>
> To check the oxygen sensor, warm up the engine and leave it running. > Disconnect the oxygen sensor electrical connection (inline with the > sensor at the exhaust manifold), and measure the output voltage of the > sensor using the DC millivolt scale of your voltmeter with respect to > ground (engine block). What you are doing with this procedure is > essentially measuring the open-loop basic CO. Assuming that the basic > CO adjustment of your engine is nearly correct, and the oxygen sensor > is working, you should see the OXS voltage between 300mV to 1V or so. > If you have a CO tester you can verify this. >
> Now, try grounding the end of the disconnected OXS wire that leads to > the electronic control unit, fooling the control unit to believe that > the engine is running real lean (which will in reality make the engine > run real rich (the control unit tries to compensate for the "lean" > condition). You should hear the idle RPM rise, and the sensor voltage > should also rise dramatically. This step shows you that the sensor > voltage is responding to changes in fuel-air mixture. If the sensor > voltage does not rise you can pretty much assume that the sensor is > dead. On the other hand, if after grounding the wire the engine RPM > does not rise, there is probably a malfunction elsewhere in the > system.
>
> You can also simulate a rich condition by connecting a 3V battery (two > AA batteries in series) to the same lead that you grounded for the > previous test. Make sure the negative terminal of the battery is > connected to ground (engine block). The fuel injection will attempt to > lean the mixture, the engine RPM should drop, and the > sensor voltage should also drop.
>
> Re-connect the oxygen sensor, and note that the sensor voltage now > varies up and down. This is closed-loop action. If the voltage > doesn't fluctuate, something is probably wrong. >
>
> -Ti
> ------------------------------------------------------------------------------

> Ti Kan                                                                  \\\
> vorsprung durch technik!                                                 \\\
> Internet: ti@altos.com                                                /// \\\
> UUCP: ...!{sun|sco|pyramid|amdahl|uunet}!altos!ti                    ////////\
>
>
>
10/09/94 Comment on note from:

From: msirota@ee.rochester.edu (Mark Sirota)

MS>On A1 cars, and on most Audis, the "button" is at the top left corner of the MS>instrument cluster, set way back from the front panel. It's a bitch, much MS>unlike the A2 cars.

Correction - Most A1 models use the same box as described above except for US made Rabbits from 81-84 they used the style mounted in the instrument cluster as Mark describes. Also on non-Lambda models, 75-80, they had an EGR light counter box that just like the OXY light is in line with the speedo cable. I believe it came on every 15k miles instead of 30 as on the Lambda OXY equipped models.

          @@ Tim Hildabrand @@  
          @@ Second Generation BBS
          @@ Headquarters for Audi and H20 VW Owners
          @@ (408) 980-8830 (2 lines) 24 hours



 

Newsgroups: rec.autos.vw
Subject: Re: [W] O2 Sensor in G60
Date: 20 Jan 1995 14:41:00 GMT
Organization: Clarkson University
Lines: 33
Bulent Murtezaoglu (mucit@cs.rochester.edu) wrote: : In article <3fjb3f$e4i@library.erc.clarkson.edu> jet@sun.soe.clarkson.edu (Jeff Thompson) writes: : [...]

: >        Question:  Can I use a 1-wire sensor when I replace my O2 sensor ??
: >               Will I only sacrifice "efficiency" at staru-up, but 
: >               keep the same engine managment functions after warm-up ??
: [...]

: As far as I know the one wire sensor is mounted on the manifold where the : exhaust gasses are hotter. A sensor mounted further down will either : take longer to warm up to the proper temperature or will never reach it. : How hot does this thing need to get to work properly? Does it just : slow down a bit when cold?

From the Bentley book on Bosch injection I seem to remember that the O2 sensor has to reach about 500F in order to work. The heater is supposed to warm it up faster, but I don't think that it heats the sensor continuously. The one-wire may be mounted closer in order to heat up faster.

Another thing I was thinking of was when/if I ever have to replace my downpipe I would get one welded up for an O2 sensor, but leave the one in the cat as well. Then I could hook a digital volt meter to the O2 sensor in the manifold and check stoichiometry when the system is running open loop (full throttle and high rpm). That way if I ever get around to doing some serious mods I will know if the stock system, Digifant II (yech), can keep up.

Later,
Jeff jet@sun.soe.clarkson.edu
: cheers

: -M

From usc!news.cerf.net!bengal.oxy.edu!acsc.com!newsserver.sdsc.edu!news.tc.cornell.edu!news.cac.psu.edu!howland.reston.ans.net!torn!nott!cunews!freenet.carleton.ca!FreeNet.Carleton.CA!bh321 Wed Feb 8 15:53:39 PST 1995 Article: 50635 of rec.autos.vw
Newsgroups: rec.autos.vw
Path: usc!news.cerf.net!bengal.oxy.edu!acsc.com!newsserver.sdsc.edu!news.tc.cornell.edu!news.cac.psu.edu!howland.reston.ans.net!torn!nott!cunews!freenet.carleton.ca!FreeNet.Carleton.CA!bh321 From: bh321@FreeNet.Carleton.CA (John V. Cianci) Subject: Re: [W] O2 Sensors (was Re: [W] O2 Sensor in G60) Message-ID: <D349y4.JBr@freenet.carleton.ca> Sender: bh321@freenet.carleton.ca (John V. Cianci) Organization: The National Capital FreeNet, Ottawa, Ontario, Canada Date: Sat, 28 Jan 1995 13:10:52 GMT
Lines: 19

O2 sensors don't try to reach a hot temperature to BURN of deposits!!! An effective O2 sensor try to heat up to a specific temperature such as 300 celsius, to operate effectively. A good sign of a inoperative O2 sensor is poor fuel consumption, or a very rough idle. The O2 sensor may still be okay, but the heater might short out within the O2 sensor, which causes a rough idle. O2 sensors DOES NOT sense fuel mixture, it gets and O2 reading in the exhaust manifold.

        A Rich mixture would read high voltage at the O2 sensor
        A Lean mixture would read low voltage at the O2 sensor
Use a multimeter could assist your in determining the state of the O2 voltage.
So an O2 does not want to burn off deposits, it wants to measure and oxygen content in the exhaust system.
--
John V. Cianci
1991 Jetta GTX 2.0L 16V
"No support group necessary, because I'm my own" A2 Rule!!!



Subject: FYI.... Changing OXY sensor...
Sender: owner-corrado-l@teleport.com
To: corrado-l@teleport.com
Reply-To: corrado-l@teleport.com
Message-Id: <01HONT3XB2KY00HTS5@utrcgw.utc.com> X-Envelope-To: JAN@FSHPP1
X-Vms-To: UTRCGW::IN%"corrado-l@teleport.com" Mime-Version: 1.0
Content-Transfer-Encoding: 7BIT
Precedence: bulk
Status: RO

For those who are interested:

I changed my oxygen sensor in my '90 G60 this past weekend. It was a VERY simple job which took me about 15 minutes to do. I probably spent more time searching for a good price on the sensor than I did installing it.
I found a variety of prices out there, so be careful. The dealer wanted $150.00. Wolfsport & others wanted $129.00. My local autoparts store wanted $139.00. I ended up purchasing it at GPR (see EC mag for number) for $98.00!!!! It was a price I couldn't refuse. It is the original Bosch 3 wire sensor. The guy (Allen) at GPR told me that I could also install a 2 wire sensor which is much cheaper ($65.00) but I didn't want to splice wires.......
I also bought the required wrench for $18.00 at GPR. It just snaps on you socket wrench.

On Sunday, I rolled my car up on ramps and searched for the sensor. It's mounted on the upper forward side of the converter. I was worried about it being 'frozen' in there, so before I did anything, I shot it with WD-40. After about 15min, I slide the wrench over the nut and pulled..... All I heard was a little snap and it was loose! That simple! (thank you anti-seize)
So then I disconnected the wiring harness which is held stationary at the rear motor mount, and pulled out the old unit. Installation was equally uneventful. The new unit even came pre-lubed with anti-sieze. Those guys at Bosch are swell!

If anyone is thinking of changing the oxy sensor, I'd recommend doing it youself. It's a piece of cake!

Pete K.

From corrado-l-owner@teleport.com Wed Mar 29 05:12 PST 1995 Received: from mailer by fshpp1 with SMTP

(1.38.193.4/16.2) id AA28972; Wed, 29 Mar 1995 05:12:48 -0800 Return-Path: <corrado-l-owner@teleport.com> Received: from desiree.teleport.com by UG.EDS.COM (PMDF V4.3-10 #4) id <01HOP2YR95Q8006I3I@UG.EDS.COM>; Wed, 29 Mar 1995 05:10:47 -0700 (PDT) Received: (from daemon@localhost) by desiree.teleport.com (8.6.10/8.6.9) id EAA02878 for corrado-l-outgoing; Wed, 29 Mar 1995 04:56:57 -0800 Received: from utrcgw.utc.com (utrcgw.utc.com [192.65.177.2]) by desiree.teleport.com (8.6.10/8.6.9) with SMTP id EAA02873 for <corrado-l@teleport.COM>; Wed, 29 Mar 1995 04:56:52 -0800 Received: from decnet-mail (C79597@MOTHER) by utrcgw.utc.com (PMDF #2906 ) id <01HOP7FCPFVK00I6X1@utrcgw.utc.com>; Wed, 29 Mar 1995 07:56:19 EST Date: Wed, 29 Mar 1995 07:56:18 -0500 (EST) From: "P. KUMMER" <C79597%MOTHER@utrcgw.utc.com> Subject: Re: FYI.....OXY Sensor
Sender: owner-corrado-l@teleport.com
To: corrado-l@teleport.com
Reply-To: corrado-l@teleport.com
Message-Id: <01HOP7FCPPJ600I6X1@utrcgw.utc.com> X-Envelope-To: JAN@FSHPP1
X-Vms-To: UTRCGW::IN%"corrado-l@teleport.com" Mime-Version: 1.0
Content-Transfer-Encoding: 7BIT
Precedence: bulk
Status: RO

>> I changed my oxygen sensor in my '90 G60 this past weekend.

>GREAT. I looked at mine last weekend...and wondered how the hell >I'd get it loose. I tried using a regular 22 mm open ended wrench >but could not fit it there, and started contemplating dropping the cat. >So that silly socket will do the trick.

I really think you need the socket to do this job & keep your sanity!

>How did you maneuver you socket over it though?

The socket has a slot in it which you can slide the wires thru. It's kinda like a crow foot with a major offset. All I did was pull the heat sheild forward from the sensor to allow me to slide the socket over the wires. Then I just pushed it onto the sensor. I negotiated my socket wrench on the right side of the cat, but upon further inspection, it seems there is much more room to the left. When I put the new one in, I tightened it on the left side because it allowed me to swing the wrench a couple of times before repositioning the socket.

When you go to remove the wire harness, I'd suggest taking the whole mounting plate off. It's held in with one bolt. This will allow you to knock the plug off the plate on a vise or similar. There was a lot of corrosion on my plate.

Good luck.

Pete
 


From ti@apple!cindy.Altos.COM Wed May 2 02:24:49 1990
Date: Mon, 30 Apr 90 20:05:38 PDT
From: ti@apple!cindy.Altos.COM (Ti Kan) Message-Id: <9005010305.AA04443@cindy.Altos.COM> To: info-vw@ames.arc.nasa.gov
Subject: Testing oxygen sensors
Status: RO

I have a good way to test the oxygen sensor to share with everyone. Since a new sensor is not cheap, you'll probably want to know that it is really dead before investing time and money on a new one.

This home-brew test requires only a DC voltmeter, and two AA batteries. This assumes the Bosch K-Jetronic with Lambda or the KE-Jetronic (virtually every fuel-injected, watercooled VW and Audi from 1980-81) system.

To check the oxygen sensor, warm up the engine and leave it running. Disconnect the oxygen sensor electrical connection (inline with the sensor at the exhaust manifold), and measure the output voltage of the sensor using the DC millivolt scale of your voltmeter with respect to ground (engine block). What you are doing with this procedure is essentially measuring the open-loop basic CO. Assuming that the basic CO adjustment of your engine is nearly correct, and the oxygen sensor is working, you should see the OXS voltage between 300mV to 1V or so. If you have a CO tester you can verify this.

Now, try grounding the end of the disconnected OXS wire that leads to the electronic control unit, fooling the control unit to believe that the engine is running real lean (which will in reality make the engine run real rich (the control unit tries to compensate for the "lean" condition). You should hear the idle RPM rise, and the sensor voltage should also rise dramatically. This step shows you that the sensor voltage is responding to changes in fuel-air mixture. If the sensor voltage does not rise you can pretty much assume that the sensor is dead. On the other hand, if after grounding the wire the engine RPM does not rise, there is probably a malfunction elsewhere in the system.

You can also simulate a rich condition by connecting a 3V battery (two AA batteries in series) to the same lead that you grounded for the previous test. Make sure the negative terminal of the battery is connected to ground (engine block). The fuel injection will attempt to lean the mixture, the engine RPM should drop, and the sensor voltage should also drop.

Re-connect the oxygen sensor, and note that the sensor voltage now varies up and down. This is closed-loop action. If the voltage doesn't fluctuate, something is probably wrong.

-Ti


Ti Kan                                                                  \\\
vorsprung durch technik!                                                 \\\
Internet: ti@altos.com                                                /// \\\
UUCP: ...!{sun|sco|pyramid|amdahl|uunet}!altos!ti                    ////////\



From: Pete Kummer <S31207%MOTHER@utrcgw.utc.com>
Subject: Re: O2 Sensor Replacement
Sender: owner-corrado-l@teleport.com
X-Vms-To: @CORRADO
Mime-Version: 1.0
Content-Transfer-Encoding: 7BIT
Precedence: bulk
Status: RO

>Ok Guys,
> I have received my replacement O2 sensor from ND (thanx!) and now I >just need some tips on where the sucker is located. I opened the hood >the other day trying to find it and had no luck. I know it is supposed >to be in the exhaust manifold, so that narrows it down to the rear of the >engine. But, there is still a lot of s*it in the way. :)

> Any tips?

>BTW: When I called ND and asked for a replacement O2 sensor, they said >that the original one was no longer made by Bosch (or something to that >effect). So, in the past they have modified the cars wiring harnes to >retrofit a different Bosch sensor. Does this sound right? I'm not >doubting TurboTim, but I just dont want to modify my baby... :(

>~~~~~~~~~~~~~~~~~~~~~~~~~T~~~~~~~~~~T~~~~~~~~~~~~~~~~~T~~~~~~~~~~/~/~Only~~~~~

>Later...                 |    \/    |   Windows95 =   |     __  / / Amiga 
>  rhurt@thepoint.net     |   \/\/   |   Macintosh84   |     \ \/ / Makes It
>                         | '90 G-60 |                 |      \/_/  Possible
I replaced my O2 sensor a few months ago. It's located on the catylitic (sp?) converter right behind the exhaust manifold. You have to get under the car to get to it. Also, if faces upward, and to the front. When I did mine, I had purchased the special wrench to do the job. It's an offset 22mm (I believe) socket. It cost me $18 bucks. The wire harness is accessible from the top.

About the different sensors, I know the Corrado uses a 5-wire O2 sensor. I was told that it can be modified to use a 3-wire sensor. The 3-wire sensor (a few months ago) was around $70. I bought the original 5-wire sensor for $98 (good deal IMO) from GPR. best price I was quoted elsewhere was $129. It was a Bosch too.

One more thing, if the sensor you purchased does not have anti-seize on the threads, then I would strongly recommend using some. It makes your next replacement much easier.

Good luck!

If you want to borrow my wrench, I'll gladly send it to you. You pay shipping!

Pete Kummer
s31207%mother@utrcgw.utc.com

From owner-corrado-l@teleport.com Mon Mar 4 13:37 PST 1996 Received: from mailer by fshpp1 with SMTP

(1.38.193.4/16.2) id AA08448; Mon, 4 Mar 1996 13:37:46 -0800 Return-Path: <owner-corrado-l@teleport.com> Received: from desiree.teleport.com by UG.EDS.COM (PMDF V4.3-10 #4) id <01I1XY0HZZZK0020CL@UG.EDS.COM>; Mon, 04 Mar 1996 13:36:34 -0700 (PDT) Received: from localhost (daemon@localhost) by desiree.teleport.com (8.6.12/8.6.9) with SMTP id NAA26410; Mon, 4 Mar 1996 13:31:48 -0800
Received: by desiree.teleport.com (bulk_mailer v1.3); Mon, 4 Mar 1996 13:31:18 -0800
Received: (from daemon@localhost) by desiree.teleport.com (8.6.12/8.6.9) id NAA26017 for corrado-l-outgoing; Mon, 4 Mar 1996 13:31:11 -0800 Received: from knights.cc.ucf.edu (knights.cc.ucf.edu [132.170.246.40]) by desiree.teleport.com (8.6.12/8.6.9) with SMTP id NAA25807 for <corrado-l@teleport.com>; Mon, 4 Mar 1996 13:30:53 -0800 Received: from Pegasus.cc.ucf.edu by knights.cc.ucf.edu with smtp RFC1413_id jrw60831 (Smail3.1.29.1 #53) id m0tthms-0006xHC; Mon, 4 Mar 96 16:27 EST Date: Mon, 04 Mar 1996 16:28:56 -0500 (EST) From: "Jonathan R. Watts" <jrw60831@pegasus.cc.ucf.edu> Subject: Oxy sensor replaced for $49. Here's how. Sender: owner-corrado-l@teleport.com
X-Sender: jrw60831@Pegasus
To: Corrado-l <corrado-l@teleport.com>
Reply-To: "Jonathan R. Watts" <jrw60831@pegasus.cc.ucf.edu> Message-Id: <Pine.SOL.3.91.960304162610.12030A-100000@Pegasus> X-Envelope-To: JAN@FSHPP1
Mime-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII Content-Transfer-Encoding: 7BIT
Precedence: bulk
Status: RO

I just saw this on rec.autos.vw.watercooled and thought it would be of interest since the oxy sensor subject comes up often around here. I can not verify the accuracy of its contents.

Jon     '90     
        53k mi.         
        Still looking for lexol leather conditioner

From: "Jonathan R. Watts" <jrw60831@pegasus.cc.ucf.edu>
Subject: Oxy sensor replaced for $49. Here's how. (fwd)
Sender: owner-corrado-l@teleport.com
X-Sender: jrw60831@Pegasus
To: Corrado-l <corrado-l@teleport.com>
Mime-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII Content-Transfer-Encoding: 7BIT
Precedence: bulk
Status: RO

I'm reposting this for the person interested in his oxy sensor. Remember, this was for a JETTA! I have no idea if the same is true for the corrado. I posted it to show the alternative method. If you have your C's o2 sensor Bosch part number, you might be able to do something similar to this person. Good luck. 


HI Gang,
I just replaced the Oxy sensor on my Jetta for $49 thanks to a friendly parts-counter guy. When I raised my eyebrows at the $135 for a VW Oxy sensor he told me a way around the big-bucks (well, relatively for a student) solution. Get a sensor for a Ford which is the exact same sensor with a different pig-tail. You will have to cut, solder and use heat-shrink tubing to splice the old VW pig-tail to the new sensor but it was worth $90. Ask for Bosch part # 13913. If you still need convincing, compare the part nos on the oxy sensor itself, identical. No charge.
Terence

From - Tue Aug 11 15:37:02 1998
From: RandyWal@ix.netcom.com (Randy Walters)
Newsgroups: rec.autos.makers.vw.watercooled
Subject: Re: 91 Jetta O2 Sensor
Date: Sat, 08 Aug 1998 01:35:33 GMT
Organization: ICGNetcom
Lines: 59

mptask@nh.ultranet.com (michael) wrote:

>I remember reading a while back in this news group about using a different O2
>sensor than the BOSH. My O2 needs to be changed, 91 Jetta 8V with 171k. The
>local dealer wants $280 to changed it. A local parts store can get me a BOSH
>-OEM- for $180. The parts store mentioned that the BOSH O2 sensor is cross
>referenced with another O2 sensor, a Standard. So for only $80 I can get the
>same O2 sensor as the BOSH. The only difference is the connectors, I would
>have to splice the wires. I haven't pulled the BOSH out yet but the Standard
>has 3 wires, 1 black and 2 white. Does anyone have any experience in splicing
>in this O2 sensor? It seems like a cheap alternative over the over priced
>BOSH. Thanks mike

You can buy relatively inexpensive generic aftermarket 3-wire oxygen sensors with the same color coded wires (2 white and 1 black) that you can splice into your old sensor's harness and connector. These can be had for less than half the price of a Bosch unit and work just as well and last many years. Royze is one brand (i've had one in my GLi for 3 years and it's still perfect) and i paid $59.00 for it. I've also heard that one for certain Fords works perfectly after it's spliced into the factory pigtail and can be had for like $49.00 at local parts stores but i haven't checked it out yet. Bosch also makes a universal 3-wire O2 sensor with the part number 13913. Oh and the two white wires are interchangeable with each other :->

WARNING:

If anyone ever splices oxygen sensor wires, make damn sure you never ever let one wire touch another wire or short to ground or you'll most likely fry most of the underhood wiring loom and some under the dash. I've seen it happen firsthand a few times (although not on my cars) and it's extremely destructive. There is a yellow hot wire that is UN-FUSED and runs from the ignition switch to many various systems in the car's electrical system and it will turn into a heating element when the ox wires are shorted out together or to ground and will melt the insulation off any other wires that come in contact with it within the loom.

Do not use un-insulated crimp splices and/or heat shrink tubing on these wires, the heat from the exhaust system melts the heat shrink and allows the newly bared splices to fuse together and that yellow wire starts-a-cookin. Stagger those splices so they're not next to each other and use only fully insulated crimp splices, and watch that the wires don't fray and short on the under-tunnel heat shield. This meltdown happened to my best friend's 87 GTi while i was in the car. Fire under the dash is no fun !! If you ever come across a solid yellow wire anywhere in the loom be afraid of it, it's hot and unfused, and branches off in several directions from a soldered junction within the loom in the engine compartment, somewhere under the master cylinder area usually. Evil little yellow wire.

Randy
85 GTi
86 GLi


From: RandyWal@ix.netcom.com (Randy Walters)
Newsgroups: rec.autos.makers.vw.watercooled
Subject: Re: O2 Sensor advice needed Please!!
Date: Wed, 10 Dec 1997 14:58:30 GMT
Organization: Netcom

On Tue, 9 Dec 1997, Tornado_Red wrote:
>> My car (1991 GTI 16v) had hi's O2 sensor replaced, I used a O2 sensor of
>> a 1989 Golf that fit my car's O2 sensor, the sensor works fine but the
>> connections (the three wires) I just put some electrical tape in it,
>> yesterday I read that IF the one of the cables in the O2 sensor came in
>> contact with each other or a metal surface it will screw my car (big NO,
>> NO) so what can I use to isolate the (three cables) that withstand high
>> temperatures, any advice will be appreciated!!

Use nylon or vinyl insulated butt splices, and stagger them apart so they can never come in contact with one another. It's ok to wrap the wires with electrical tape after they're properly butt-spliced but route the wires so they're not close enough to the exhaust pipe to melt anything, and make sure they can't be pinched or shorted against any metal. It's pretty important :-)

"Karl W. Schneider" <kschneid@eecs.wsu.edu> wrote:
>Someone please correct me if I'm wrong, but I speculate that the original >
O2 sensor is a 3-wire heated unit, while the replacement may not be
>heated. The heating-unit positive ("hot") lead is unfused, according to
>the Bentley for my '86 GTI. You definitely don't want this lead shorting
>to ground, or you may melt circuitry. The heating improves performance
>after startup, because the unit needs to be at 600 deg F to operate properly.

Not may, it *will* melt circuitry (among many other things :-) I think all three-wire sensors are heated, but i dunno for sure.

>Mark at Adirondack Auto Parts, in a recent e-mail to me, "strongly
>recommends ... using shrink tubing for insulating/sealing."
>-- Karl

Well i wouldn't use heat shrink as the only means of insulation. I once used the supplied naked metal butt splices and the supplied heat shrink tubing but after a few weeks the heat from the nearby exhaust pipe eventually over-shrunk the heat-shrink which then split open and allowed two of the connections to touch each other. The car would not stay running when cold, it would die after a few seconds. Repeated starts would eventually warm up the ox sensor sufficiently and the car would suddenly run perfect, as if someone threw a switch. It had me totally baffled until i followed the cold troubleshooting procedures for the oxygen sensor control unit as outlined in the Bentley Manual and discovered i had continuity between terminals 2 & 8 where there should be _none_ when cold. Closer inspection of the sensor loom revealed the problem at my splices. I pulled them apart and the problem disappeared! Needless to say i switched to nylon insulated butt splices and also staggered them so there'd be no chance of them ever coming together again, then wrapped the whole mess with high quality 33+ electrical tape. I'm lucky that the un-fused yellow hotwire wasn't one of the shorted wires or the whole loom would've most likely been destroyed, and the car may have even burned down. The under dash wiring on my best friend's 87 GTi caught on fire within a few miles of driving with his new Euro downpipe. He'd allowed the relocated oxygen sensor wires to fray on the tunnel shield and it totally fried the whole loom under the hood and under the dash. Not a pretty sight ! There are a few other stories of melted looms due to ox sensor wire shorting that i'm personally aware of.

Everybody open your Bentley Manual to the Electronic Engine Control wiring diagram and find the O2 sensor, then follow it's yellow wire through the diagram and see how many electronic components it branches off to. This is the wire that becomes an un-fused heating element that takes out all the wiring within the looms. That's one scary wire !

Randy
85 GTi
86 GLi