Possible bad O2 sensor?

[AirCooledCurtis]

Looking for some ideas on what to check. Here's the scenario:

'86 Digifant Syncro. It was apart for a few months getting too much work done. While apart it got the injectors cleaned and flow tested, plugs and wires, TPS adjusted and checked at the plug on the manifold, o2 signal wire replaced (Vanagain kit), fuel lines, fuel filter, fuel pump. AFM physically bench tested and cleaned.

On startup, cold or hot, the engine will immediately fire and find idle. It'll rev well and respond to throttle input exactly as it should. After it gets warm, or if hot start, 10 seconds or so has passed, it'll progressively stop wanting to rev until the engine will just die when you give it throttle. The o2 shows it go super super lean with throttle. While idling (it idles well) the o2 swings from .1-1V at a healthy tempo. Engine off key on I get .5V at the o2 signal wire when unplugged (good). Pin 19 at the ECU plug to the chassis ground under the back seat shows that the ground reference wire is good.

If I unplug the o2 signal wire this issue goes away and the van will rev.

I've had a similar issue with the o2 sensor before on another '86, I'm just puzzled why the o2 is causing this and what else I can rule out.

[DigiMatrix]

Do you have anything connected to the O2 signal wire other than the ECU? I had an issue when I had an O2 led monitor connected to my O2 signal and it ran fine for a long time and then caused me all sorts of grief. When I finally ripped it out in frustration, the issue was gone. Lesson to self - never connect a secondary gauge to the O2 signal wire. Weld in a second bung, add a second O2 sensor and use that for whst ever you want.
_________________
_________________
'91 Syncro Westfalia
soon to be 2.5 liter stroker WBX 10:1 comp
Digifant ECU with knock sensor and custom
cylinder selective knock control by me

[AirCooledCurtis]

[zerotofifty]

If in doubt, change it out. oxygen senders are consumable, a bad one can ruin a very expensive catalyst.

[AirCooledCurtis]

[latelogan]

i just want to remind you to check the ground wires to the o2 sensor as well as the signal wire and power wire. That caused my shop and me about 3 weeks of heck. Although, your symptoms sound different than mine (black exhaust on start).

ohm them to the frame is prob best.


Wondering if this is more like the temp sensor. I had a bad connection on mine and replaced the entire connector.
_________________
1991 Westy stock 2.1 auto transmission
1978 Westy (Sold)
1971 Westy (Sold)

[crazyvwvanman]

Pin 19 doesn't reference the chassis ground, it references the engine ground while the engine is running.

The wiring bundle that connects to the hall sender also has a brown wire that doesn't go on to the hall, but is supposed to be screwed down to bare engine metal.
That wire connects to pin 19 to ground the shield and provide the O2 sensor reference ground for use by the ECU.

Due to the alternator output currents which vary by what loads are being powered, there can be small and varying voltage differences between the chassis metal and the engine metal.
Since the O2 sensor signal works with fractions of a volt and relies on its threaded contact with the exhaust for the signal ground, it needs to use a good engine ground to avoid having its reading affected by the alternator currents.

Edit: to perhaps clarify, the pin 19 should not have a direct wire connection to chassis ground and didn't from the factory by design.

Mark

[AirCooledCurtis]

[AirCooledCurtis]

[crazyvwvanman]

As another check, try to measure on low DC volts with a meter probe on the shield strands near the sensor and the other on bare engine metal.
Of course the engine should be running and the alternator warning light should have turned off.

Mark

[DanHoug]

a poor ground on the alternator can raise the voltage potential of the engine block and can mess up O2 signal to the ECU. so many have added an extra ground from the alternator case to the body.

but more likely in your scenario is an exhaust leak introducing oxygen that the O2 sensor sees as a lean mixture and consequently over-richens the mixture to the point where it runs terrible. additionally, anything that causes a cylinder to miss adds oxygen to the exhaust stream and can cause a low voltage on the sensor.

and just to make sure, when you say it goes "super lean" you are seeing a LOW voltage on the O2 sensor, correct?
_________________
-dan
60% of what you find on the internet is wrong, including this post.

'87 Westy & '89 Westy both 2.1 4spd

Past projects can be found at--
www.thefixitworkshop.com

[1988M5]

[AirCooledCurtis]

[AirCooledCurtis]

[zerotofifty]

[DigiMatrix]

[AirCooledCurtis]

[AirCooledCurtis]

An update for people wanting to play armchair mechanic. I installed the new Bosch o2, and it's still doing the same thing, except it's not better if I unplug the o2. Not sure what was happening before but I can't replicate it now. So I turned to looking to other systems. The ECU is seeing the TPS. The idle control valve isn't doing anything, I jumped the 2 wires on the PS pump which is supposed to raise the idle, right? Well it didn't do anything. I pulled the top off of the AFM, and if I physically richen the mixture while it's stalling, which isn't really scientific or probably healthy, it'll not stall. I have to nearly max out the arm so it needs a lot of encouragement to add fuel. I clamped the regulator return and the issue really improves. I don't have a way to test the fuel pressure right now, but the fuel pump and filter, along with lines, are new so I trust that the pressure is at least higher than what the regulator is suppose to "regulate" It's having a grossly under fuel situation that for some reason somewhat goes away with a key cycle.

My flat lifter pumped up so it's not clacking, so I can confidently say that it's not got any exhaust leaks. If it does it's a slow seep and won't be causing mixture issues.

[AirCooledCurtis]

More updates, the fuel pressure is good. 38psi with the vacuum line unhooked, 32psi with some vacuum. It also holds pressure. I'm thinking that the same buffoon who hacked up the o2 signal wire could've messed with the AFM. It's a rebuilt Syncro one and the traces look good. I'll check that next, I'm not ruling out the ECU though.

Could the idle control circuit be causing any issue off idle?

[AirCooledCurtis]

This morning I tested output voltage from the AFM while the car was off but key position 2. No dead areas. I tried to replicate sudden movements in the range that it'll die and I couldn't get anything out of the ordinary. I may have to test it while it's on the car and running. It still seems to me like the lambda control in the ECU is toast.