2.0 FSI Poor idle when cold

[saabman]

Heres the Trim values v RPM you can clearly see as soon as the o2 sensor values are applied to the engine starts running right

[JustCruisn]

So is this saying the O2 heaters are not working ?

[MEL744]

stored and used as the start up value but is not right for the condition of the engine at that time

This sounds very reasonable. Seems to be using some fixed value, would potentially explain why it runs fine if you restart it - it calculates a new value or maybe the heaters have more time to work?

Apparently clearing engine fault codes resets the fuel trim values - I wonder if clearing the codes prior to a cold start makes a difference?

I reset a check engine light the other day - I don't recall it making a difference but will retry it any pay more attention.

Unfortunatley I cant try this out as my daughter lives 200km away so I can only work on the data I have

Let me know if there are any values you'd think would be particularly interesting to log and I'll get them

So is this saying the O2 heaters are not working ?

Maybe, but they might be working fine, I'm not entirely sure how quickly they should heat up. You also seem to have the 2.0 FSI, do you have this rough start condition?

[MEL744]

Made a few plots related to the Lambda values today. This was from a cold start and I drove it until it came up to temperature on the dash - however nothing much interesting happens after 5 mins so I truncated the data there. The vertical red line represents when the rough idle period ends. Plot is available here (recommend downloading and viewing on your computer as Google Drive viewer isn't the best).

[saabman]

That looks great.
A picture certainly tells a thousand words ...

The readings from Bank 1 seem to be all over the shop while bank 2 is reasonably consistent . On mine its bank 2 that is all over the place and 1 is more constant. is it the sensor or is it something else

I come across this tonight which might help.

http://www.vaglinks.com/Docs/Audi/B5/S4/BentleyPublisher.com_Audi_B5_S4_O2-Sensor_Checking.pdf

[MEL744]

That link looks excellent.

A couple of observations I made:

1) The high idle lasted for longer on this start than previously. Normally it lasts 30-40 seconds, yesterday night it was 75 seconds. The car had been sitting for 48 hours at that point.

2) Bank 1 Sensor 1 not only displays erratic behaviour much sooner than the others - it also seems to oscillate around 1.5 V rather than 0.5V. Is it dodgy or is the label file wrong and this is some other sensor?

3) Sometimes there is a terrible rattle during the high idle - could it be related? Sticking lifter? Cam chain rattle?

Based on the abnormal start (extended high idle) I will reproduce the graphs today and see if that changes anything. Normally the misfires do not occur immediately but start to happen about 10 seconds in. I want to see if there is any correlation there. Additionally from that link it appears as though the Lambda Control % should not exceed +/- 4% during idle. Bank 2 sensor 1 seems to stay within this limit however bank 1 sensor 1 clearly exceeds it.

[MEL744]

Made some more graphs today. Decided to swap out MB 043 for MB 014 since nothing much was happening on 43 and 14 has the misfire counter. Added a green line to signify when the first misfire is detected (could be misfiring before this but misfire detection seems to be off then). Still the voltage on B1S1 is definitely the most odd thing that's jumping out at me. Also made a plot of a normal 'hot' start, but did not let it idle before setting off so keep that in mind when viewing. Was going to plot them on top of each other but it just made it harder to decipher due to the oscillating nature of the data.

Cold start

Hot start

[saabman]

So is this saying the O2 heaters are not working ?

I dont think so it seems to be the time it takes to get going

[saabman]

Just looking at your readings are almost identical to mine - the engine load is a good one makes it look like your developing 70 odd KW just to idle but its just because the manifold vacuum is so low. As I understand it, the manifold vacuum is low because the timing is so far retarded youll see its about 25deg ATDC. I have taken the misfires as a result of such late ignition timing, a symptom rather than a cause.

The rattle might be a clue but then it might be just a result of the misfires?

How accesible are the sensors - I wonder if its easy enough to switch them over and see if the erratic voltage changes banks. if it changes then the sensor is the cause if it doesnt then there is something else.

I think the label file is good enough. I think the errtaic readings are real - I dont have logs for the voltage from my sensors but the lambda values are definantly diferent between the two.

[MEL744]

Yep I agree with all of that. Was going to try swapping the sensors but I am not sure where they are. Will have to do some research and give it a go.

[JustCruisn]

Maybe, but they might be working fine, I'm not entirely sure how quickly they should heat up. You also seem to have the 2.0 FSI, do you have this rough start condition?

Yep I have the same. Can be rough idle, engine kind of surges or feels unbalanced. Reversing out of driveway required a lot of brake effort to keep it from racing off.
After about 30 seconds the idle drops off & the engine runs silky smooth.

[MEL744]

Yep I have the same. Can be rough idle, engine kind of surges or feels unbalanced. Reversing out of driveway required a lot of brake effort to keep it from racing off.
After about 30 seconds the idle drops off & the engine runs silky smooth.

Very interesting - seems to be an extremely widespread issue.

I was having a look around today in the engine bay and at some stuff online. Looks like the pre-cat sensor sits on top of the manifold and is easy to access (maybe need to remove some EGR piping to get it out). Naturally the post-cat sensor is going to be under the car but from some tutorials (on GTIs) it looks easy to remove once you lift the car. Haven't got the time at the moment to swap them over, hopefully I will be able to try it in the next few weeks.

[saabman]

What’s your engine? Mine is BLX

I thought you also had 2 sensors on the manifold.

Yes just about every scrap of information is either for the diesel or the GTI the lowly 2.0FSI seems to be somewhat unpopular

[MEL744]

Yeah my bad, I have a BVY with 4 sensors. I can't see any others except this one so far from a quick check. From that site you linked earlier the GTI does in fact have 2 sensors only - why the 2.0 FSI needs 4 baffles me.

EDIT: Went back and did some more research

Only one is visible from above (the one I photographed) - this is Bank 1 Sensor 1 (Lambda Probe G39)

The other 3 sensors are all after the cat according to the workshop manual site

Exhaust manifold with B1S1

The rest of the exhaust system with 3 sensors

[saabman]

Yes I looked into it as we’ll I have BLX engine and it also has 4 sensors but 2 in the manifold and 1 after the cat and think the other one is after the NOX cat (to be confirmed)

I assumed (incorrectly) the BVY also had 2 in the manifold.

So my idea of swapping the sensors in the manifold won’t work for you.

They don’t make it easy.

My current thinking is along the lines of, you have 4 sensors all in a line the readings should be all be at least very similar. if there is a discrepancy the issue is either between two sensors eg leaking exhaust, damaged cat. Or one of the sensors is faulty. ( does the BVY have air injection? The BLX doesn’t)

This is really a good exercise in getting to know the ins and outs of our engines and the subtle differences between them.