P2098 and CEL

B6/7 S4/RS4
#1

I’ve recently been getting a P2098 code which triggers the CEL. I scanned it and cleared it with a cheapo code reader I keep in the glove but after 250 miles it has come back on. After looking at the live data on the reader I noticed my long term fuel trims are out of whack as well. Bank 1 was around 9% and bank 2 was 0%. O2 sensors have about 30k miles on them and the readings seemed to be within the parameters that were given in the (cheap) scan tool. What’s the best course for resolving these issues? Do you think they are related?

Car has 111k miles on it, has been running great and just got checked over for vacuum leaks.

Here’s my findings with VCDS today.

008344 - Post-Catalyst Oxygen (Lambda) Sensor Correction; Bank 2
P2098 - 002 - Too Lean - MIL ON
Freeze Frame:
Fault Status: 11100010
Fault Priority: 0
Fault Frequency: 2
Reset counter: 255
Mileage: 179126 km
Time Indication: 0

         Freeze Frame:
                RPM: 2156 /min
                Load: 11.4 %
                Speed: 69.0 km/h
                Temperature: 90.0°C
                Temperature: 5.0°C
                Absolute Pres.: 1010.0 mbar
                Voltage: 13.843 V

Not sure if these were the right groups to log but they are a couple 3rd gear pulls.

https://docs.google.com/spreadsheets/d/1Flc4LslRRSmFkKrpZVQ69J88xc17v5atQKnNKkjynLo/edit?usp=sharing

16:39:10 Group 003: General
760 /min Engine Speed (G28)
6.64 g/s Mass Air Flow Sensor (G70)
0.8 % Throttle Valve Angle
6.0 °BTDC Ignition Timing Angle

16:39:10 Group 002: General
760 /min Engine Speed (G28)
18.8 % Engine Load
3.28 ms Injection Timing
6.69 g/s Mass Air Flow Sensor (G70)

16:39:10 Group 032: Lambda Control (Mixture Adaptation)
-2.6 % Adaptation (Idle) Bank 1 Sensor 1
9.4 % Adaptation (Partial) Bank 1 Sensor 1
0.0 % Adaptation (Idle) Bank 2 Sensor 1
0.0 % Adaptation (Partial) Bank 2 Sensor 1

#2

Check that your 5mm intake manifold bolts are tight. Have you taken the engine air box apart at all recently? Sometimes the ECU is tricky about turbulence from a filter being crooked as weird as that sounds.

#3

9% is pretty lean… That means the computer is adding 9% more fuel to the engine to bring the fuel trims back to the 14.7:1 ratio. If it is localized to one bank (bank 2 in this case) then i would start poking around that bank to see if the intake manifold bolts have worked loose(as mentioned by jimmy bones) very common and can cause leaks on on bank. can you do any type of data logging with that scanner? like see what happens to the MAF as the engine RPM’s increase? I cant remember if these engines have a MAP sensor. That could give you good data to try and diagnose the issue. A dirty MAF can make an engine run lean, but its usually both banks. You could always do the propane test…

#4

Jim

Good job getting the codes and the data up. When you look at it you will see that bank 2 is the side that your getting the code for and you will see that bank two also shows no activity. The lean code would possibly mean the sensor is dead and the car is correcting off the main 02. Check the fuel trims for the LTFT for bank two and see if it adjusted or reads. If its still at ZERO that would mean the 02 is more linely bad.

#5

i totally missed the sensor not giving data…my bad…before you run out and change the o2 sensor…do you have access to a scope? you should check to make sure there is voltage at the sensor first. i’ve seen more than one that ended up being an open wire or a short to ground.

#6

It would be a good idea to check the measuring value blocks for the front O2s. Try out blocks 33-35.

#7

Thank you all for the replies!

That was the first thing I checked after the CEL came on the first time. A couple bolts needed to be re-torqued but weren’t completely loose. Car was smoke checked and there doesn’t seem to be any vac leaks. I cleaned my air filter about 10k miles ago and that was the last time it was touched.

This all seems to make sense. Where do I check the LTFT’s in VCDS? In the description of the code it says “Post catalyst O2 sensor” wouldn’t that be sensor 2? But you’re suggesting its probably the front sensor that is bad, correct?

I’ll get some logs at lunch today. I’m sure most of you have seen this on the ross tech site but here’s some good info on the measuring blocks in VCDS.

http://www.ross-tech.com/vag-com/m_blocks/030-049.html

#8

https://docs.google.com/spreadsheets/d/1zELvLp8ZLh_EyVzBvknvcw0Ssx_ZGQmTiKHGA4eQdV8

Logged groups 033 - 035 on my drive home from work. Logging stopped for some reason 100 seconds in but I restarted it in my driveway and it looks like the test for bank 2 sensor 1 never ran. Safe to assume the sensor should be replaced at this point?

#9

Try one last thing before you replace the sensor.

Get the engine warmed up to the middle of the gauge (at least 80 degrees C coolant temp), find a good parking spot out of the way and put the shifter in neutral/out of gear, engage the parking brake, hook up your vagcom, go into 01-engine electronics, basic settings, channel 35, click the icon to turn the test on, put your left foot on the brake pedal, floor the gas pedal with your right foot, the rpms should raise up to around 2K rpms and hold there. If the rpms fly up past 3K rpms then stop. Wait for the O2 sensor test to come on and run. Let us know if it passes or doesn’t.

That is the Audi scan tool procedure for testing the O2s but the JHM tune changes certain tests so that may not work the way that I describe it.

I personally like to replace the front O2s in pairs but that is just me. The front O2s are rather expensive though.

#10

The car went into the test mode as you said but the actual test never started running on the bank 2 sensor, just said “TEST OFF” as rpms held at 1600. Tried it on the bank 1 sensor after clearing and the test started almost immediately and got the B1S1 - OK message after about 15 seconds.

My local NAPA had one Bosch 17353 sensor in stock and price matched to amazon for $96. I’ll report back after the sensor goes in.

#11

Sensor went in last week and the code was cleared. Then checked the bank 2 O2 aging test in basic settings and it ran and passed. Yay. Bank 1 passed as well. LTFT’s however are still off and reading close to what they were before the sensor was replaced. Bank 1 ~10% and bank 2 ~ 1%, they’ve been fluctuating slightly over the last couple days. This morning CEL came back on and its the same P2098 code.

What’s my next step?

#12

Ok to recap you actually had a bad 02 and now thats fixed. But your still getting a code. Did you get the freeze frame data for the code. if you can post that that will be helpful

#13

Yes. As far as I could tell bank 2 sensor 1 was no good. I attempted to run the aging test on the old sensor as Jimmy recommended and it would not go into test mode. After it was replaced it immediately went into the test mode and then gave me the “B2-S1 OK” message.

Just found another issue. I was pulling into the shop to do an oil change and the exhaust sounded a little strange when I got close to a wall. Got it up on the lift and found an exhaust leak on the driver’s side at the flex pipe, which is after the first sensor and before the second.

So is the lack of exhaust gasses getting to the second sensor causing the lean code? But why would this mess the fuel trims on bank 1 so much?

1 Fault Found:

008344 - Post-Catalyst Oxygen (Lambda) Sensor Correction; Bank 2
P2098 - 002 - Too Lean - MIL ON
Freeze Frame:
Fault Status: 11100010
Fault Priority: 0
Fault Frequency: 1
Reset counter: 255
Mileage: 179728 km
Time Indication: 0

         Freeze Frame:
                RPM: 1954 /min
                Load: 18.4 %
                Speed: 62.0 km/h
                Temperature: 90.0°C
                Temperature: 12.0°C
                Absolute Pres.: 1010.0 mbar
                Voltage: 13.335 V
#14

1 Fault Found:

008344 - Post-Catalyst Oxygen (Lambda) Sensor Correction; Bank 2
P2098 - 002 - Too Lean - MIL ON

This code typically means ends up being a faulty rear o2 sensor.

#15

So lets say I replace the rear O2, it fixes the CEL and my fuel trims are still reading like they are now. Should I be concerned with that?

#16

In my opinion +10% is a little high for LTFT, but it is within the normal range. Ideally it would be close to 0%.

Did you fix the exhaust leak? That could cause the lean code on bank 2.

The exhaust leak on bank 2 wouldn’t have any effect on bank 1’s FT readings. You have another problem somewhere else on bank 1.

Start with checking for vacuum leaks and exhaust leaks on the passenger side. You could also have a fueling issue specific to bank 1 that is causing the high LTFT reading.

Can you rerun the logs you did before, but include group 33 this time? Log 10 seconds each at idle, 1500rpm, and 2500rpm.

#17

Thanks for the reply. I did a temporary fix on the exhaust leak last night and ran a quick log on the fuel trims. Nothing changed. Car was scheduled to bring to my mechanic this morning so I’m hoping they can figure it out, I just don’t have the time to mess with it. I’m hoping the exhaust leak was what was triggering the code. When this first started it appeared the rear o2’s were functioning properly. When I cleared the code in the past it has taken about 250 miles to come back on so its probably going to be a drive it and see what happens situation with the CEL.

For the problem with the fuel trims I don’t think its a vacuum leak. I just checked it over for that and there were none. I also checked for other exhaust leaks last night when I patched up the flex pipe. None that I could see.

If my mechanic can’t figure it out I’ll run the logs when I get the car back.

#18

Your fuel trims wont get effected that quick. If you log the instant fuel trims that can give you an idea but you really need to get the car under good load to see a vac leak.

  • or - 10 isnt too bad but it also depends on what the other bank is at. In a ideal siuation you would want both banks within 5% of eachother.

To check LTFT you would want to clear the codes wait for the 02s to warm up and then drive the car in heavy load at low RPM.

#19

After rebuilding my leaky downpipes the issue has been resolved.

I had no idea how bad of shape the pipes were in until they were removed. 50k miles and 5 New England winters took a serious toll. The flex pipes were completely shot, O2 sensor bungs were porous around the welds and the flanges were slightly bent at the manifold from being over tightened. After sand blasting, two $15 flex pipe sections, some JB weld on the bungs and high temp primer and paint they’re as good as new.

After driving for about a week the LTFT’s are staying steady at ~5% for bank 1 and ~-2% for bank 2. Not within 5% of each other but its better than before and the car is running great.


http://i1304.photobucket.com/albums/s527/jimmys4/Mobile%20Uploads/image_zpsho7admzr.jpeg


http://i1304.photobucket.com/albums/s527/jimmys4/Mobile%20Uploads/image_zpstrhqw8pz.jpeg


http://i1304.photobucket.com/albums/s527/jimmys4/Mobile%20Uploads/image_zpspwb5ff3d.jpeg


http://i1304.photobucket.com/albums/s527/jimmys4/Mobile%20Uploads/image_zpsylpzlm1m.jpeg

#20

Wow that was in bad shape. Something that bad off will cause ltft issues. Nice fix and update