E39 M54B25 Mysterious lean issues

[seronen]

Hello guys,

(I'm from Finland so English grammar is not my strong suit)

I have a e39 525i (M54B25) and it has some strange issues. First of all, there is one fault code, P0171, Lean on bank 1. No other codes.
Looking at livedata, the short term fuel trim for Bank 1 is like +26.7% which sound absolutely crazy. (And the bank1 sensor voltage is around 0.0015V) So I replaced the lambda with no change.
Intake manifold checked, all vacuum lines and intake boots replaced, plugs behind the manifold replaced. CCV replaced. All gaskets in the intake side replaced. After no fix or even improvement to the issue, I started to look at the engine and the exhaust. Spark plugs were all very black, so it was running rich even on bank1. Then I realised what if it is a leak in the exhaust manifold before the upstream lambda sensor, and there was some leakage. (Good old soap water and a blower trick) And the exhaust manifolds were at a shop, they said exhaust gasket on bank1 was leaking and they replaced both exhaust manifold gaskets.
The problem is still present. Immediately when the motor goes into closed loop, the stft on bank1 goes sky high. Bank2 is relatively unaffected by the issue, stft around 0 and ltft 0. Ltft on bank1 impossible to extract because the ecu goes into some sort of failsafe when realising the lambda on b1 "doesn't work properly". The car has the "Euro2" tune which eliminates both downstream lambdas and possibly sap (don't know for sure)

I've had this issue for months now and I'm starting to run out of money and patience!

Any ideas guys?

[JimLev]

Welcome to the forum Seronen.
Your English is perfect, much better than some of the peeps here in the US.
About the only info I can suggest is 3 bad fuel injectors (probably not), pinched or shorted O2 wire, and to get a smoke test done.
Good luck.

[seronen]

Welcome to the forum Seronen.
Your English is perfect, much better than some of the peeps here in the US.
About the only info I can suggest is 3 bad fuel injectors (probably not), pinched or shorted O2 wire, and to get a smoke test done.
Good luck.

Thanks!
I have switched the injector locations between banks, so I'd rule that one out. Wires have been measured by a pro and me, so I doubt that but I gotta check them out once again so I can sleep well. And yea, smoke test would be a good thing to do. Especially on the exhaust side. I still think the exhaust is leaking somehow. The lambda reads so lean. It's just the placement of the exhaust manifolds that drives me crazy!

Thanks for your insight JimLev.

[RosieE39]

Have you actually measured the O2 sensor voltage at the DME with a multimeter or scope?

[seronen]

Have you actually measured the O2 sensor voltage at the DME with a multimeter or scope?

I have measured the wires for resistance and measured the actual output of the lambda. The output was equal to what livedata said. (0.0005V)

[RosieE39]

I have measured the wires for resistance and measured the actual output of the lambda. The output was equal to what livedata said. (0.0005V)

Have you done voltage drop tests between Pins 4 - 14 and 3 - 20?

[seronen]

Have you done voltage drop tests between Pins 4 - 14 and 3 - 20?

Nah I haven't done voltage tests on the pins. But I checked the resistance between those pins on the ecu and the lambda connector. I assume the wiring is ok as the resistance is within normal limits. And I checked the voltage output straight from the lambda sensors connector when the car was running. Being the same what the livedata tells me I ruled out the problem caused by bad wiring. And the sleeve covering the wires to the lambda is in mint condition, but I know that doesn't mean that the wiring inside is good....

[seronen]

I doubt this is an electrical issue, because now when the exhaust manifold gaskets have been replaced, the voltage isn't dropping all the way to 0. I followed it from livedata for a while and it's in the range of 0.015V-0.020V (Sorry for the numbers being incorrect before, there is one 0 too much) But the lean condition improved very slightly. But there is a downside, now the bank1 won't go into the "safety factory injection maps", so it's running very rich
All this started when I was driving as usual, and suddenly the engine started misfiring A LOT and stalled on me. I traced that problem to the gas tank, drivers side fuel jet pump broken or loose, not transferring fuel to the fuel pump side. Added gasoline and away I go. And then the lambda sensor is going totally insane. I'm afraid that the cylinder head is cracked from the exhaust side on bank1. Can't see, can't confirm.

I just hope this won't be a wallet draining, bank balance annihilator- type of issue. It's really starting to sound like one.

[RosieE39]

Very pertinent information.
Do a smoke test on your inlet.
The inlet manifold probably has a crack on one of the first 3 runners because of a “backfire” when it ran out of fuel.

[seronen]

Very pertinent information.
Do a smoke test on your inlet.
The inlet manifold probably has a crack on one of the first 3 runners because of a “backfire” when it ran out of fuel.

Allright, will do as soon as the sun rises here. That pretty much could be the case, as the problem is bank relevant. Thanks!

[seronen]

Very pertinent information.
Do a smoke test on your inlet.
The inlet manifold probably has a crack on one of the first 3 runners because of a “backfire” when it ran out of fuel.

Doing a smoke test here is like 90€ if you want a proper test. So I took the manifold off and visually inspected it. No cracks or anything, it's in mint condition. There was a joint line which looked strange so I applied some tape over it without any changes. All was good.

I noted also that on bank2 the idle stft is about 0, but now the ltft is about +6.5. So the problem seems to affecting greatly on bank1 and when driving it's affecting bank2 also. Sounds like a classical vacuum leak but where on earth is it??

Also, found a hairline crack on my valve cover on the exhaust side, around cylinders 4-5, close to the screw mount. It's leaking oil. Can that cause the problem?

[RosieE39]

The valve cover leak isn’t the cause of your problem.

Here’s a very cheap and easy homemade smoke machine:-
https://www.instructables.com/id/BES...You-Can-Build/

[danb1979]

When was the last time you looked at the DISA and the O-ring that is sandwiched between it and the inlet manifold?

If that's disappeared, it'll be dragging in extra air that way too, after the MAF, giving you issues

Worth looking into, even if to just rule it out

[91calysorot]

Check/clean your Mass Air Flow sensor (MAF) could be dirty or not functioning correctly.

[seronen]

When was the last time you looked at the DISA and the O-ring that is sandwiched between it and the inlet manifold?

If that's disappeared, it'll be dragging in extra air that way too, after the MAF, giving you issues

Worth looking into, even if to just rule it out

Disa replaced with a oem bmw part in december. The o-ring in the new part is nice and tight.

- - - Updated - - -

Check/clean your Mass Air Flow sensor (MAF) could be dirty or not functioning correctly.

MAF cleaned with proper poison, and the data it's giving is text book. Though I have only been following it at idle. But I will do a test with a different MAF. I'll borrow one from a friend.

[seronen]

A quick update, no luck with the smoke test, though it was a bit sketchy... Next up, I'm going to measure the exhaust back pressure from the pre-cat lambda sensor holes. It's not the first time that when bank2 cat is clogged it causes a lean code on bank1. I assume the pressure is getting away from bank2 via the SAP path to bank1. This is my best guess. I'll update when it's done.

But there is no misfires even on high rpm, so I really cant tell for sure that the cat is going. And also no data from the post-cat O2 sensors due to the euro2 tune.

[seronen]

Allright, back pressure tested and it was ok for both banks. Close to zero psi even on high rpm range. I don't really know what do do next.
I ran the engine without the MAF, and the bank2 stft went crazy rich, like -22% and bank1 stft was solid lean +27.5%. Is this a sign for a failing MAF?

Also, when ripping the intake manifold off, i noticed in one of the the intake manifold runners there was no plastic "plug" on the smaller hole under the intake manifold gasket which aligns with the head. I was wondering whats the purpose of these smaller holes? 60931-691923df5be458836c37010046f98cb3.jpg

[1823alex]

Hey I had this exact issue I lost or didn't have one of the little plugs as well so I pulled some from the junk yard. I had a very minor lean condition but I don't think the plug changed anything as it's all already sealed regardless.

Sent from my Pixel 3 XL using Tapatalk

[RosieE39]

I was wondering whats the purpose of these smaller holes? 60931-691923df5be458836c37010046f98cb3.jpg

They’re the air passages for the idle circuit.

[seronen]

I have some new data which is blowing my mind completely.

Out of nowhere, the bank 1 is doing its job correctly. And now, bank 2 is the one with the lean issue. The o2 sensor on bank 2 was stuck lean, so I replaced it with the old one from bank 1. No changes. I will probably go and buy a new one so I can be sure that the sensor is operating like it should.

How can this be possible? All I did to the car was just to play around with the o2 sensors. I didn't even switch their locations or anything.

However, after this I switched the o2 sensors between banks and voila, the lean condition followed. So the lambda is gone.

And for the cherry on top, TPS is stuck at 9.5% at any given rpm, even when WOT. So new o2 sensor and throttle body i guess.

I should get paid for this This diagnosing is turning into a full-time job!

!! UPDATE !!

Ok the car is trash.
Yesterday the issue was on bank 2 so I switched to o2 sensor locations between banks. The condition followed, so i bought a new o2 sensor. So now, the o2 sensor which yesterday worked correctly is in bank 2 and the new one is in bank 1. Started the car, runs fine, stft for both banks around 0. I drove around, getting the system heated up. Came back home and plugged my computer in. Bank 1 OK, bank 2 stft 26.6 in idle. WHAT!? So now the lambda which worked yesterday is failing?

I think the next step is to tear out to main wiring harness from the car and inspecting it very, very closely.

[seronen]

Time for a guessing game guys...

The o2's have been replaced. I ran the car today and gathered data on different conditions. So when I start the car, it runs perfect. The short term fuel trims are very good, bouncing around -3% to +3%. Then suddenly the bank 2 stft goes rocketing up, stopping at +26.6%. This issue can be reproduced. The only variable is the time between starting the car and when the stft spikes up. Sometimes it's 5min and sometimes 15min. But it goes there and doesn't come down. And the stft rise is so fast, like it was closing injectors or something like that. It takes only about 2-3 sec for it to go from 0% to 26.6%

So the real question here is that what can cause such an abnormal issue? Is my DME is fried? If the DME f.ex. had to shut down injectors, it would tell it to me with a code, right? There is no codes, other than P0174 (Bank 2 system too lean)
I don't trust the lean condition at all. I have driven the car with the engine "lean" and when I pulled the o2's they were black, not leanish colored.

There was a SIB for a matter of this type. O2 control error of some sorts, fixed by re-flashing the DME. But I don't want to take the car to the dealer because $$$$$, unless you demand me to. Taking your car to a shop here in Finland is not cheap...

Thanks for your time in advance.

[1823alex]

Time for a guessing game guys...

The o2's have been replaced. I ran the car today and gathered data on different conditions. So when I start the car, it runs perfect. The short term fuel trims are very good, bouncing around -3% to +3%. Then suddenly the bank 2 stft goes rocketing up, stopping at +26.6%. This issue can be reproduced. The only variable is the time between starting the car and when the stft spikes up. Sometimes it's 5min and sometimes 15min. But it goes there and doesn't come down. And the stft rise is so fast, like it was closing injectors or something like that. It takes only about 2-3 sec for it to go from 0% to 26.6%

So the real question here is that what can cause such an abnormal issue? Is my DME is fried? If the DME f.ex. had to shut down injectors, it would tell it to me with a code, right? There is no codes, other than P0174 (Bank 2 system too lean)
I don't trust the lean condition at all. I have driven the car with the engine "lean" and when I pulled the o2's they were black, not leanish colored.

There was a SIB for a matter of this type. O2 control error of some sorts, fixed by re-flashing the DME. But I don't want to take the car to the dealer because $$$$$, unless you demand me to. Taking your car to a shop here in Finland is not cheap...

Thanks for your time in advance.

The codes are quite confusing. Using an actual BMW scanner may help you but usually the p0174 I had, had a BMW code definition saying that the car was running lean but it couldn't richen the air fuel mixture anymore.

Sent from my Pixel 3 XL using Tapatalk

[seronen]

The codes are quite confusing. Using an actual BMW scanner may help you but usually the p0174 I had, had a BMW code definition saying that the car was running lean but it couldn't richen the air fuel mixture anymore.

Sent from my Pixel 3 XL using Tapatalk

I don't yet have INPA, it's a ongoing project...
Yeah you are right. But I understand the concept of engine being lean. Usually too much air due to intake leak. Stft's are the correction % that the DME has to make to the injection to maintain the combustion stoichiometrically correct.
And lean codes are triggered when the DME can't correct the mixture anymore, yes. (well a bit earlier I suppose)
This is an odd situation because I can't take the codes seriously. Yes the stft shows the bank is lean, but is it really? In this case it probably isn't due to visual inspection of the cylinder and it's parts. I earlier pulled the spark plugs on the "lean" bank and they were all pitch black as well as the pre cat O2 on that bank. If the bank is truly running so lean that the DME can't correct it, it should not be black. That is a sign that the bank is running very rich. Please correct me if I'm wrong.
And the fact that the stft rises sky high in a heartbeat out of nowhere, is supporting my theory if you ask me.

Sorry for the long posts, I usually tend to get a bit carried away

[1823alex]

I don't yet have INPA, it's a ongoing project...
Yeah you are right. But I understand the concept of engine being lean. Usually too much air due to intake leak. Stft's are the correction % that the DME has to make to the injection to maintain the combustion stoichiometrically correct.
And lean codes are triggered when the DME can't correct the mixture anymore, yes. (well a bit earlier I suppose)
This is an odd situation because I can't take the codes seriously. Yes the stft shows the bank is lean, but is it really? In this case it probably isn't due to visual inspection of the cylinder and it's parts. I earlier pulled the spark plugs on the "lean" bank and they were all pitch black as well as the pre cat O2 on that bank. If the bank is truly running so lean that the DME can't correct it, it should not be black. That is a sign that the bank is running very rich. Please correct me if I'm wrong.
And the fact that the stft rises sky high in a heartbeat out of nowhere, is supporting my theory if you ask me.

Sorry for the long posts, I usually tend to get a bit carried away

It IS running very rich, but that's because it's adding more and more fuel to the mixture just to get it to run. It is running reach due to the fact that there is a vacuum leak.

You have a vacuum leak somewhere most likely.

Check the intake boots that are after the MAF and also the tiny vacuum line hoses behind and under the intake manifold.

Sent from my Pixel 3 XL using Tapatalk

[seronen]

It IS running very rich, but that's because it's adding more and more fuel to the mixture just to get it to run. It is running reach due to the fact that there is a vacuum leak.

You have a vacuum leak somewhere most likely.

Check the intake boots that are after the MAF and also the tiny vacuum line hoses behind and under the intake manifold.

Sent from my Pixel 3 XL using Tapatalk

Intake leak not possible. Intake checked 3 times and ALL rubber parts and hoses and gaskets replaced. And smoke tested to be intact.

If there was a vacuum leak, it would not run rich. Of course it's adding fuel, but there would be also more air because the system is getting air what the MAF doesn't know of.
If your stft is let's say +10%, it isn't running rich. It's just adding 10% more fuel to keep the AFR correct. The AFR is still correct 14.7 parts of air and 1 part of fuel.
And when the code is set, the DME can't compensate the excess air anymore and then the engine will run lean due to the lack of fuel burnt. (Stft around +26%)