temperature dependent O2 sensor
 

2000 A6 4.2, 115000 miles. Problem is with bank 2 sensor 1 intermittent. I assume it is a wiring issue, but i can't find it. I'm reluctant to probe the engine controller connector, i have seen the contacts damaged from careless probing.

I had a code for intermittent sensor, so i replaced the sensor with the Bosch OEM part.

Now i get P0154 code after the coolant temperature gets near 90C.

Watching the lambda values with the Ross-Tech tool, bank 1 ramps between ~0% and -5%. Bank 2 ramps between ~0% and +5%. When the coolant temperature gets near 90C, bank 2 sensor 1 goes to +25% and the voltage to ~50 mV.

i cleaned the connectors.

No amount of pulling, pushing, or bending on the wire harness or connectors will make the sensor fail at low temperature or work after it has been hot. Letting the car cool overnight starts the process all over. The sensor wire is not near the exhaust pipe

I recently had the transmission out and i have a leak in the exhaust at the resonator, which i was hoping to hold off replacing until next summer.

i did find the passenger side (US drive) intake manifold flap stuck and fixed that.

No codes indicating air leaks.

If i drive it until it is hot, i get codes for misfire on 5, 6, 7, and 8, and rough running.

No other codes.

i assume it is wiring, the harness passes behind the coolant expansion tank, but i'm not having success tracking it down.

My question after all of that is: does something clever or special happen when the coolant temperature gets to ~90C?

Yes, the engine temp at 90c or 194 degrees defines "open loop to closed loop" so the ECU can adjust fuel trim and other controlled engine management outputs.

No other codes? I'd be checking for vacuum leaks too.

I would probe the 02 sensor connection at the firewall...
 

with a DVOM…you should see rapid crossovers between essentially zero/.1V and what, 0.9V?

An oxygen sensor will typically generate up to about 0.9 volts when the fuel mixture is rich and there is little unburned oxygen in the exhaust. When the mixture is lean, the sensor's output voltage will drop down to about 0.1 volts.

If you don't see this crossover and it should be at least once per second or faster, either the sensor is bad or the ECU isn't getting the information and changing the mixture.

Here's one source:
http://densoautoparts.com/sites/defa...ing%20tips.pdf

Um, I was answering the OP's question about the CTS reaching operating temp and what the ECU does at that point, who are you replying too?

Regarding O2 sensors the heaters and the SAI compensate for the cold start along with the ECU knowing that the CTS is under 90c "open loop" some fuel enrichment and timing advance is added until the operating temp is met then the "closed loop" events take place, we should mention what occurs from cold start idles, warm idles and above idles during RPM loads.

I have the Ross-Tech tool. Plotting the lambda % shows these transitions at ~1 sec rate at low temperature. Above 90C, bank 1 continues this, but bank 2 ramps up to 25% and stays there.

When i bought the car in 2010 with 100,000 miles on it i replaced all of the 3mm vacuum lines in the engine compartment.

Apparently when i changed the bank 2 sensor 1 i disturbed things enough to create a leak. I will drop the electrical checks for now and look for leaks

Changing that sensor was a royal pita, it is cocked ~30 degrees towards the center, where if it was located vertical it would come right out, almost straight up, after moving the coolant expansion tank.

Audi should make their engineers repair some of their cars.

"...the engine temp at 90c or 194 degrees defines "open loop to closed loop"..."

Thanks, that is what i needed to know.

A simple question always deserves a simple answer, no problem ;)

This is still not fixed. i have searched for vacuum leaks using a MiteyVac and vacuum gauge. I checked the fuel evap canistor purge valve, it opens and closes. The secondary air combination valves do close. If it was vacuum leak, why does this issue only show up on the bank two O2 sensor?

I made an adapter to connect my Volt meter to the sensor leads at the ECU connector. Sensor voltage fluctuates between ~ 0.1 Volt and 0.75 Volt while cold. After driving around to get coo.lant temperature above 87 C, the voltage still fluctuates, but over a small range around 0.1 Volt. When i turned off the engine, the sensor voltage quickly rose to ~ 0.45 Volt. That tells me that the sensor and wiring are OK, and something is causing the ECU to reduce fuel to the left bank above 87C.

I am thinking possibly i disturbed the left bank fuel rail while is was effing around trying to repair the manifold flapper, and some of the injector seals are leaking.

Here are the codes i had shortly after i changed bank 2 sensor one, and drove the car until it stumbled:
16534 - Oxygen (Lambda) Sensor B2 S1
P0150 - 35-00 - Malfunction in Circuit
17546 - Fuel Trim: Bank 2 (Add)
P1138 - 35-00 - System too Lean
17538 - Fuel Trim; Bank 2 (Mult)
P1130 - 35-00 - System too Lean
16806 - Main Catalyst; Bank 1
P0422 - 35-00 - Efficiency Below Threshold
16684 - Random/Multiple Cylinder Misfire Detected
P0300 - 35-00 -
16692 - Cylinder 8
P0308 - 35-00 - Misfire Detected
16691 - Cylinder 7
P0307 - 35-00 - Misfire Detected
16689 - Cylinder 5
P0305 - 35-00 - Misfire Detected
16690 - Cylinder 6
P0306 - 35-00 - Misfire Detected

No codes showed up today, but i only drove it until it got hot.
Here are the lambda readings after the coolant got above 87C:
13:31:03 Group 001
680 /min RPM
87.0°C Temperature
2.3 % Lambda
0.0 % Lambda

13:31:03 Group 033
0.0 % Lambda
0.680 V Voltage
0.0 % Lambda
0.095 V Voltage

13:31:03 Group 031
0.700 V Voltage
0.070 V Voltage
0.095 V Voltage
0.840 V Voltage

Real time lambda value on bank 2 sensor 1 was stuck at 25%.

Any ideas?

NOt to be rude, but sometimes you just have to take it to a competent audi technician
 

.