2001 A4 1.8t AWM - -25% Lambda at idle - runs rich - no codes

[mobildetroit]

I have a 2001 A4 1.8t AWM Quattro Avant. I've been having a really hard time diagnosing this one.

Runs rich -25% Lambda on group 107 when idle.
Using Ross-Tech VCDS for diagnostic.
While driving it's all over the place but stays in the negatives most of the time, sometimes at 0 sometimes within spec, often out of spec on the rich side.
When I first start the car it's on 0% and then slowly descends down to -25%

Is it true that a leak on the pressure side of the intake could cause the engine to run this rich?
I would think, especially at idle, that the engine would run lean, not rich.
I don't seem to have any obvious vacuum leaks, everything is where it should be.

Car seems low on power at low RPM. Turbo does engage, have not done any kind of boost level test, not sure what stock boost should be on this car or if that would be an accurate test.
I've been trying to figure this one out for awhile and have not been able to.

All the parts that have been changed on the car recently (with no change in rich condition):
MAF
Intake air temp
Coolant temp
Plugs and coils
Engine speed
Upstream O2 sensor
Cat to Turbo gasket
Suction pump valve assembly (hoses and check valves)
PCV valve

I did a test on the secondary air system, I did get an Abort result...could that be a factor? If so why?

I bought this car used and as far as I know the engine computer has a factory configuration, but I wonder sometimes if it's got a tune on it and that's why it's not running right.
Not sure what to do.

Anybody have any ideas or experience with this issue?

[mobildetroit]

I discovered something today. When I take the oil fill cap off the Lambda values fall right into the middle. So there seems to be a connection to how the crankcase vents. Anybody have any ideas how crank case ventilation could cause a rich mixture?

[walky_talky20]

Post your long term trims (block 032) and post what you see for a MAF value at idle (when it's -25%)

Then watch your MAF value with and without the oil cap on.

For it to affect it that much, I would think there is some modification of the PCV breather system. Like venting the valve cover or hooking something up wrong. If there is nothing obvious you can watch your lambda trim or MAF value while you pinch off various hoses in the engine bay. Like the vacuum supply line to the Suction Jet Pump, for one. Or the line that connects the SJP to the PCV valve.

Also, I would make sure there aren't any modifications to fueling. Different size MAF housing, different fuel pressure regulator, different injectors than stock, etc.

[mobildetroit]

Injectors are new and stock type.
MAF is new and stock type.
Fuel pressure regulator is new and stock 4bar type.
Pressure holds in rail so injectors are not leaking down and regulator is not leaking back.
Suction jet pump valve system assembly was recently replaced.
block 32 is -1.5% and 0% for first two values respectively.
No change with oil cap off
No change when pinching manifold to suction jet pump.
Goes down to -6.6% on idle bank 1 sensor 1 when pinching between PCV and suction jet pump (I don't know what this means).

Vacuum at idle is 15 in. Hg vac. Shouldn't it be near 20?
I guess this suggests a vacuum leak but I don't understand why a vacuum leak would cause a rich running condition...

[mobildetroit]

The -25% didn't change at all for any of these tests.
However, earlier today I had the oil cap off and it changed to something higher then -25% but eventually was back at -25%.
Did the oil cap test again after your post at it didn't change from -25%...but I'll venture to say it could change but not sure what changes it.
It's consistently at -25% at idle, when driving it's all over the place but typically in the -'s

MAP sensor says 850 mbar actual when 820 mbar is specified when at idle.
Took it for a drive and actual seems always lower then specified.
I live in Colorado and the altitude is about 6000ft.
BARO sensor reads correctly.

[Rodgman15]

Vacuum at idle should be 20in/hg. Or close anyway. When I start my car when it's -30 here it's 15in/hg. You've got a leak.

[mobildetroit]

I agree that it's likely a leak but I'm having trouble finding the leak.

[mobildetroit]

I disconnected all vacuum lines on intake manifold and plugged. I'm still at 15 Hg. So I don't suspect a vacuum leak now. If there is a leak it's between manifold and head, manifold and throttle body or injector seats. However, I sprayed it all down with starter fluid and the engine didn't skip a beat. So I then suspected maybe valves are not sealing, so I seafoamed it. Hg did come up a few notches to about 17~18, so I'll likely seafoam again soon. During the seafoam, that's when I saw smoke coming from where the exhaust manifold meets the head. So at this point I'm thinking I have loose nuts on the exhaust manifold and the leak is causing a lean condition on the upstream O2 sensor, thus causing the computer to run the engine rich to compensate....problem is, there isn't enough O2 on the intake side and thus the fuel doesn't burn completely making the problem worse....in other words an extreme -25% lambda. So in other words the exhaust leak may be creating a sort of feed back loop....this may explain why when I first start the car it's at 0% lambda and then slowly descends down to -25% (cold or hot), then just stays there when at idle.

Letting it cool down then I'm going to see if I have loose nuts on the exhaust manifold.

[mobildetroit]

Manifold nuts fine, nice and tight. Turbo is bolted on nice and tight...no signs of gasket blowout, etc...
I tried opening up the exhaust between turbo and cat to see if maybe cat back pressure was causing low vacuum, but vacuum remained at about 15 Hg and still at -25% lambda at idle...
Took valve cover off and cam chain timing looks good...but noticed something interesting on the cam belt pulley.
This car was previously owned by someone else and not sure what kind of work was done on the car.
Paint mark on cam timing belt pulley is off a bit. I suspect someone used the bolt to turn the engine over and tweeked it a bit, but how do you know for sure or if the cam sheered it's key a bit?

[walky_talky20]

The check for camshaft keyway (without removing the t-belt) is as follows:
- Line up the front camshaft gear TDC mark to the valve cover. Make sure it's "on the money".
- Don't move anything
- Pop the valve cover off
- Check that the notch at the rear of the exhaust camshaft is lined up directly with it's arrow.
- If they don't match up, check the keyway. Otherwise it's OK.


Thoughts:
- I think your timing is fine.
- Your Long Term Trims (block 032) seem perfect. Does the -25% trim also show in block 001?
- Do you have other evidence of running rich? Sooty plugs, dark exhaust, terrible mpg etc?
- Have you verified that your Coolant Temp Sensor is reading correctly (reasonable values) in the ECU?
- [Theory: CTS reads "cold" and never warms up = Cold start is OK, but runs increasingly rich as engine temp increases and the CTS reading gets further and further from the truth.]
- Have you looked at the "Target AFR vs Actual AFR" in VCDS block 031?
- I'd like to see a datalog of group 031 as the car warms up (watch it along with group 107, to see it's reaction as it goes from 0% to -25%). Then another log driving (light cruise), and one more log at hard accel (3rd gear pull, wide open throttle from 1500 to 5k RPM). Do that last log one more time, but logging the MAF instead. (This is just a suggestion for some data to collect. I'm not telling you what to do).

I know you changed the MAF and CTS, but new parts are not always perfect. And there can be issues with the wiring to the sensor, and not actually the sensor itself. Those 2 sensors are the ones with the most power over fuel quantity. With a 25% trim, I'm most suspect of the data from those 2 sensors.

[mobildetroit]

I took the cam pulley bolt out and the key looks fine.
I'm going to put the valve cover back on today, bring cyl 1 up to TDC and check the crank timing belt pulley to verify the crank is on the mark along with the cam on the mark.
I suspect if it's off a belt tooth that could cause a slight retard and drop in vacuum. I'm currently at 15 Hg during idle along with the -25% lambda.
If I had a problem with fuel trim related sensors wouldn't my vacuum be closer to normal regardless?

It's very very much running rich. Can smell gas in the exhaust and all plugs are black/wet.
Car lacks power at low RPM. Seems to use lots of fuel in short period of time but I haven't really tried to measure fuel consumption.
I don't remember what 031 says, I guess I can look at it once I get the car back together, but it really doesn't answer to the vacuum issue, does it?
CTS reads correctly.

[mobildetroit]

If there were wiring issues to the MAF I would likely get engine computer error codes about circuit faults.
If there were wiring issues to the CTS I would likely get a wrong reading for temp or error codes about circuit faults.
Am I wrong about that?

[walky_talky20]

One would hope. But it is not always the case.

If the circuit is entirely severed (open circuit), or if the circuit wires are touching where they shouldn't be (short circuit), then you might get a code for that. But if there is excessive resistance in a wire to the CTS, you may not get a code. The CTS works on resistance, so extra resistance just adds to that of the sensor - offsetting the final value by some amount. Same with the MAF - it outputs a voltage, but if the wire has too much resistance - some voltage will be dropped across the wire, giving a reading that is very much still present - yet incorrect.

[Joe Jr.]

Not trying to beat a dead horse....but i traced down some vacuum/boost leaks on mine over the weekend and have been through it before on my 01 and my 00. My experience between the 2 cars has been usually from a vacuum or boost leak...hopefully its that simple for you also....

[mobildetroit]

were you getting -25% lambda when you had your leaks.
My timing is good, all the marks line up. Even checked crank belt pulley.

MAF and CTS read correct values according to the computer.

[vrmm]

Have you taken your altitude into account for this 15hg? I never saw above 18. Evvver. It was usually 16-17. 20 is in theory where it should be but I could never hit it even after a billion rebuilds, new intake piping, etcetc.

[mobildetroit]

I verified the BARO sensor reading is correct. I was suspecting altitude being a factor as well, I'm about 6000 ft above sea level.
At this point, because the lambda value is way off, I'm suspecting a bad ECU (or maybe the previous owner tried modifying it and broke the maps).
It's always kind of ran rich.

I noticed the O2 sensor wires frayed last year so I insulated them with high temp RTV. The temp sensor wires were missing insulation at the base where they go into the connector, so they could have shorted a long time ago, and I also re-insulated those wires. I've replaced the O2 sensor so the bad wire there was fixed.
The temp sensor reads correctly according to the ECU, however, the heater on the upstream O2 sensor seems to keep cycling between on and off over and over again, not sure if that's normal.
Does the upstream O2 sensor cycle the sensor heater on and off over and over again while in normal running mode on your cars?
I'm wondering if the O2 sensor shorted and caused some circuit damage in the ECU, for example, fried a surface mount resistor fusing the two pads together with 0 or creating a very high (out of expected range) resistance....I wonder if something like that could cause the computer to richen the mixture and still think there is nothing wrong.

I just ordered another ECU off Ebay, should get it in a few days.
Meanwhile I'm going to test the O2 and CTS wiring back to the computer to make sure there are no shorts.
I've got a new CTS connector coming as well.

[mobildetroit]

Replaced ECU...no change in behavior...still running rich.
I don't suspect an issue with the ECU.

At this point the only thing left to resolve is the fuel pressure question.
Fuel pump is new, fuel pressure regulator is new, injectors are new.....fuel pressure is about 10 psi to high.
WTF Audi!?!?!?

[walky_talky20]

Could be a restriction in the fuel return line to the tank. It can't dump the pressure.

[mobildetroit]

I'm down to an issue in fuel system. I suspected a restriction in return line already, put my pressure gauge on it and it never went over 2psi....but maybe that's enough to toss off the entire system.
I do know there is some piping inside of the fuel tank connected to the return line...I guess maybe there could be a slight restriction in there somewhere as well?
Like maybe the return line is filling up a chamber in the fuel tank up to the return line and causing slight blockage?
What is interesting is that when I start the car, lambda is 0% then slowly inches down to -25%.
However it always does that....so it may not be enough proof that there is a fluid based restriction in the tank.

Is the fuel tank not venting properly? Are the return lines inside of the tank not position in the correct locations?
Could the EVAP system cause fuel return line pressure increases?
I have already tried running the car with the fuel cap off to see if Lambda values go up from -25% and they do not.

Does anyone know of a diagram of the fuel line routing inside of a 2001 a4 quattro fuel tank?

Was thinking about maybe running a rubber line into a gas can off the return from the fuel rail to see if lambda results change....that should eliminate any restriction.

[walky_talky20]

Was going to suggest the rubber line into gas can.

[mobildetroit]

I confirmed there is no restriction in the fuel return line. I disconnected it and piped the return into a gas can, started the car and it went to -25% lambda like it usually does.

At idle pressure at the fuel rail is 67psi, it should be around 50psi.

The FPR was replaced, the fuel pump was replaced. It's a VDO brand fuel pump and a BWD brand FPR.
The FPR is a 4 bar regulator and I've confirmed it's the correct rating for the engine.
The fuel injectors were also replaced with new (not re-manufactured) injectors.
The vacuum line between intake and FPR is new and in good shape, no leaks.

This is a 2001 A4 Quattro AWM.....

Is it possible for the engine to idle rich if the pressure is to high in the fuel rail? I believe so...
The fuel rail holds pressure when the pump is shut off, so I don't suspect a leak past the FPR.

Is there some sort of restrictor valve on the fuel filters on this year car that could be failing? I know on later models there is some sort of FPR in the fuel filter.

I'm kind of at a loss at this point...why would the fuel pressure be so high?

Or maybe the brand FPR I have in the car is garbage...?

If the car is supposed to have 50psi at the rail, why does Audi say there should be a 4 bar regulator in the car? Shouldn't it be a 3.5 bar regulator?

[mobildetroit]

Is this what should be in the car even though audi dealer said it should be a 4 bar?
https://www.bahnbrenner.com/vw_audi/...ator_New_Style

[mobildetroit]

I just ordered a new 4 bar Bosch Fuel Pressure Regulator part# 0280160575.

If that doesn't bring the pressure down I may see if I can get a 3.5 bar regulator...or maybe an adjustable.

[garytightpants]

Is this what should be in the car even though audi dealer said it should be a 4 bar?
https://www.bahnbrenner.com/vw_audi/...ator_New_Style

No. OEM are 3bar or 4bar, no 3.5 bar

[mobildetroit]

That link I posted is for a 3.5 bar regulator that is supposed to fit the 2001 A4 Quattro.....however, I'm getting a new Bosch 4 bar regulator to see if the rail pressure drops down to around 50psi like it's supposed to be. I think the other new BDW brand FPR is just poorly manufactured or is marked incorrectly regarding it's rating.

[garytightpants]

That link I posted is for a 3.5 bar regulator that is supposed to fit the 2001 A4 Quattro.....however, I'm getting a new Bosch 4 bar regulator to see if the rail pressure drops down to around 50psi like it's supposed to be. I think the other new BDW brand FPR is just poorly manufactured or is marked incorrectly regarding it's rating.

yes, that FPR will fit. My answer was in regards to you asking if it is what SHOULD be in there.

[mobildetroit]

What should be in a 2001 AWM is a 4bar FPR according to Audi.

[mobildetroit]

Not trying to beat a dead horse....but i traced down some vacuum/boost leaks on mine over the weekend and have been through it before on my 01 and my 00. My experience between the 2 cars has been usually from a vacuum or boost leak...hopefully its that simple for you also....

Just to follow up...I don't believe it's a vacuum/boost leak. The idle vacuum is not perfect but it's not enough to cause 67psi at the fuel rail. In fact I've manually pumped down vacuum on the FPR to over -20 inHg and the fuel rail pressure did not go down to the 50psi it should have been at idle. Right now the only logical thing to do is replace the FPR....again....with another new one but of a different brand.

[mobildetroit]

The check for camshaft keyway (without removing the t-belt) is as follows:
- Line up the front camshaft gear TDC mark to the valve cover. Make sure it's "on the money".
- Don't move anything
- Pop the valve cover off
- Check that the notch at the rear of the exhaust camshaft is lined up directly with it's arrow.
- If they don't match up, check the keyway. Otherwise it's OK.


Thoughts:
- I think your timing is fine.
- Your Long Term Trims (block 032) seem perfect. Does the -25% trim also show in block 001?
- Do you have other evidence of running rich? Sooty plugs, dark exhaust, terrible mpg etc?
- Have you verified that your Coolant Temp Sensor is reading correctly (reasonable values) in the ECU?
- [Theory: CTS reads "cold" and never warms up = Cold start is OK, but runs increasingly rich as engine temp increases and the CTS reading gets further and further from the truth.]
- Have you looked at the "Target AFR vs Actual AFR" in VCDS block 031?
- I'd like to see a datalog of group 031 as the car warms up (watch it along with group 107, to see it's reaction as it goes from 0% to -25%). Then another log driving (light cruise), and one more log at hard accel (3rd gear pull, wide open throttle from 1500 to 5k RPM). Do that last log one more time, but logging the MAF instead. (This is just a suggestion for some data to collect. I'm not telling you what to do).

I know you changed the MAF and CTS, but new parts are not always perfect. And there can be issues with the wiring to the sensor, and not actually the sensor itself. Those 2 sensors are the ones with the most power over fuel quantity. With a 25% trim, I'm most suspect of the data from those 2 sensors.

So what values should the MAF and CTS be at with the engine at idle and warmed up? If there is an issue with the wiring or the sensors then those values at idle should be incorrect (because I have -25% at idle)?
I don't think I need to collect data during a drive to prove if the sensor signals to the computer are the issue being that the issue is most prevalent at idle.

[walky_talky20]

At hot idle:
MAF: something like 3 or 4 g/s
CTS: approximately 90C (usually between 89 and 95C)

[mobildetroit]

All is well with the sensors. I'm starting to suspect maybe a vacuum leak at the waste gate or maybe a waste gate flapper issue. Had a pressure gauge on car while driving and it never went over 4 or 5 psi under heavy acceleration, then on first deceleration it would hit 10psi for a split second then go to 0. However, at times when vacuum was at 20 while driving the lambda mix was at 0.....but as soon as the engine goes back down to idle and vacuum goes to around 15 lambda goes to -25%.

I'm wondering if the waste gate isn't completely closed at idle and it's messing with intake vacuum. Or maybe there is a vacuum leak in the waste gate?
So I'm going to be checking this next.

[mobildetroit]

After closer inspection I can see what looks to be some rubber hose cord frays/rip in the rubber on the line connected to the waste gate. Very likely I have a slight leak there. It explains why the boost level may be low. Not sure how this could cause the high fuel pressure unless it's the reason why the car idles at about 15 vac. I think the N75 is working correctly because with the control wire disconnected it barely does 1 psi. With it connected it does 5 psi under load. So highest probability it's vacuum line to the waste gate.

[walky_talky20]

The wastegate will never see vacuum, as its connected to the compressor outlet - it will see boost only. It can leak air out, though. That would definitely cause wacky boost control issues and rich idle (it will lose metered air at idle).

[mobildetroit]

I connected my vacuum tester to the wastegate and pumped it down to check for diaphragm leaks and I had none.
With the hose between wastegate and N75 disconnected the car develops 5 psi boost and behaves the same as if the N75 is connected.
Does not appear to be any leaks in hose.

So....issue with wastegate or issue with N75?

If there is a constant leak in the N75 valve that could explain why the turbo does the same thing disconnected as connected to the N75, knowing that there are no leaks in the wastegate hose or diaphragm. Wastegate shaft moves with tension on it.

So Im suspecting the wastegate is always open to the atmosphere and when RPM goes up the wastegate spring gives out and the exhaust blows past the wastegate valve (fail safe?)
I guess I could test this by sealing off the wastegate so it's not exposed to the atmosphere, that should make it boost to the limit of the turbo.

System could be working except maybe an issue with the wastegate valve in the turbo seating fully? Maybe it's physically leaking due to carbon, burning or some other warpage?
Would that cause -25% lambda at idle?

[mobildetroit]

I sealed the line on the wastegate diaphram and no change, still boosts to almost 5psi and then stops there.
Will the turbo boost to 5 psi with a waste gate valve leak?
Does the wastegate actuator shaft/rod need an adjustment?
Why is the behavior the same with the N75 connected?
Is 5psi normal boost for a 2001 A4 Quattro Avant 1.8L AWM? I keep reading the car will be at 5psi in limp mode...but car isn't running in limp mode. So I suspect 5psi is abnormally low.

I just ordered a new N75, going to try that, although I blocked off the control line to the wastegate and it still seems to open.

I'm thinking this suggests there is an issue mechanically with the wastegate flapper or something?

[mobildetroit]

I took the N75 off the car today and put my vacuum tester on the intake side of the valve (the long tube). There was a vacuum leak on that. Based on what I've read this tube should stay sealed when there is no power to the N75. However in my case there is a leak there. If I seal the other two tubes there is no leak. So the valve plunger is leaking. This means the pressure reference line from the pressure side of the compressor housing is leaking back into the no pressure intake side through the N75 at idle.

This doesn't rule out an issue with the waste gate but could be contributing to the low intake manifold vacuum at idle and possibly messing with the FPR causing high fuel pressure. I really doubt this is what's causing the high fuel pressure, so its just a guess at the moment. I have a new N75 valve coming, I'm going to put it on, check vacuum and Lambda at idle. If it's in or closer to the normal range of -10 ~ 10% then the N75 valve was the issue.

[vrmm]

That's a lot of work put into it :p just put a 3 bar on from... A b5 2.8? I can't recall what they came on but they're common. No need to spend 60$.. find one at a junk yard for like 5$. I've done it while tuned with meastro on a 4 bar and the 3 bar dropped afr's 3 'points' or so. Can't comment on +/-%.

I have had an aftermarket external wastegate have a problem where the plunger stuck out about 2mm all the time. It didn't have an effect of afr/lambda readings, just took boost way longer to build and would never really got max boost. I took it apart and ground down the... C clip seat or something.. It's been years, but it seated correctly afterward. Given that it was external you could hear it and feel it when putting your hand on the outlet pipe. Not that that really helps... But just a little experience in the realm of what you were tracking down.

[mobildetroit]

vrmm....thanks for the comment regarding your turbo situation. I put all new vacuum lines on everything, silicone, proper type for turbo applications. I put the new N75 on. I put a boost gauge on the car. I can't get it to produce more then 5 psi boost. I vacuum tested the waste gate actuator and it held vacuum so that means the diaphragm is not leaking. The control shaft seems to move although seems stiff, likely due to spring in wastegate actuator? However, it only generates 5psi. I've pressurized the intake and it doesn't leak.

The thing that you said that is interesting is that it sounds like you experienced no effect on mixture even though your wastegate valve was stuck open.
I am/was suspecting that the wastegate valve/plunger is free wheeling...not connected to actuator rod or something. I've read somewhere that 5 psi all the time is an indication the car is running in limp mode. But I get no engine codes regarding such mode or issues that would warrant it...no codes at all generally....however since I replaced the vacuum lines I now get a running rich code. So I guess now the engine is doing a better job of knowing that it's not running correctly?

I'm thinking about going with an adjustable FPR....but I shouldn't have to do that on a stock car so I'm reluctant.

I guess I'm still down to some sort of leak in the intake system....is it possible the turbo compressor housing is cracked or leaking? Would that cause the fuel pressure regulator to allow a higher fuel rail pressure at idle?
When I drive the car the only time I hit -25% is at idle and usually when I'm stopped. When I'm coasting at a higher RPM Lambda seems to stay around 0%

[mobildetroit]

I have pinched the control hose to the wastegate actuator expecting to see boost build above 5psi and it didn't...it either made no boost or it continued to make 5psi...it was one or the other, don't remember which.

I have a feeling I'm going to have to take the turbo off to inspect it. Anyone have any advice as to what to look for regarding such inspection?