Best guess for negative fuel trim problem ?

[MrSkarZ]

Hello,

I have been chasing down a negative fuel trim for a while on a M52B28 (around 6 months) I though I solved the problem last year changing injectors, haven't drive the car much this winter, starting to drive again more recently and noticed that it is still smelling rich when park + black exhaust pipe + blackish spark plug + INPA -30 fuel trim both banks

fuel trim 0.png
fuel trim.png
after cleaning carbon from O2 sensors & few days after adaptation values reset
fuel trim netative.png

The car is a 1995 328i e36 Euro, no secondary air pump, euro M3 exhaust not leaking, injectors not leaking either and have been refurbished last year, last time that I checked (6 months ago) fuel pressure was fine and FPR was changing rpm while removing vacuum line. I plugged an old intake vacuum gauge on the brake booster line to check if I have valve problems but NTR.

Tomorrow I am going to create a vacuum leak to test O2 sensors, expecting them to read more O2 and therefore the fuel trim values should go up. I am a little skeptical that the two sensors would have failed at the same time.
I am also considering getting a new FPR for sanity

should not be a side effect of M50 manifold swap right ?

[bmwdirtracer]

Oxygen sensors on a '95 model year BMW are worn out by ~120k kilometers. Don't clean them, replace them. A fuel pressure test needs to be done while the car is driving; long term fuel trims are not measured at idle; they are measured at cruise speed.

You should also watch live data for the MAF sensor; it may be telling the computer that there is more air coming in than is really entering the engine. It's possible that an earlier. freer flowing manifold might cause missed readings. Did you follow an accepted procedure for the manifold swap, like using the recommended MAF/ throttle body?

By the way, your links didn't work for me.

[MrSkarZ]

Sorry about the pictures I renewed the attachments is it working now ? I cleaned the O2 sensors first as they came with the M3 exhaust mani 5 years ago, and looks quite good expect the carbon build up on the tip, the previous owner didn't tell me if they were old or new when he did the swap, I am going to look for replacement.

I can't really take fuel pressure while driving, I have to borrow a gauge from somewhere and it is a pain in this ass on this car, the access point are bellow the car near fuel filter, there is no schrader valve on the fuel rail, only FPR at the back near firewall. Could it work if I plug in from inside near fuel pump, bellow the rear sit ?

MAF data at idle :
- cold ~ 18-20kg/h
analogue V1.PNG
- warm 14-15kg/h
analogue V1bis.png
analogue 1.png

For the M50 intake manifold swap, I got a simple kit which use a thick flat gasket between the intake mani round gasket and throttle body round gasket, I have been thinking building a metal one but I didn't notice any leaks so far.

[dworthy]

You have a vacuum leak, I can tell you that because of the Engine Load is off the charts. That too can effect the fuel trims, so have the intake checked out using a smoke machine.

[MrSkarZ]

@Dworthy I can check again for vacuum leak in intake later today, but I though that intake vacuum leak only generate positive fuel trim as the extra O2 enter the system after MAF, then is reported by the O2 sensors and as result tell the DME to send more fuel to burn the extra O2.
As for engine load it has always been off the chart, can't find any good info on that, all M52 INPA pics I have seen are off chart for engine load...

This morning after a bit of driving, I created an artificial intake leak (removing oil cap), engine idle rough, plus there was a small suction force at the oil cap (supposed to be normal). As expected INPA reads high positive values in the fuel trims somehow the O2 sensors are working and noticed the extra O2.

After stopping the vacuum leak, idle return to normal and fuel trims to -30. Something else that I have noticed, is that since I cleaned the O2 my ticking problem seems to get less pronounced right after some driving then return to normal (louder) after idling for 5 minutes. It may not been related but I noticed it as I have been living with that damn ticking for already 4 years without any success solving or interfering with it (oil, lifter, tensionner, chain, exhaust crack/leak ...).

As Bmwdirtracer suggest, the MAF is maybe crazy and report more air than the reality or O2 sensors are failing. What is normal idle kg/h value for MAF ?
Is there also a way to check for clog catalyst converter ?

[dworthy]

Anything in the Green zone would be normal on INPA. Anytime I see an "!" that means for you to look at that area specifically. So have a look at your MAF and ISV to ensure both are clean and passes the bench test.

When you take your oil cap off, that is just confirming that the CCV is working properly, and not a way to find vacuum leaks. Trust me with the engine load off the charts like that always pointed to a vacuum leak for me.

[bmwdirtracer]

Did '95 Euro 328's really have 5 volt wide band O2 sensors? (Darin? What do you say?) We definitely had only 1 volt sensors here, for a while past that.

I don't use INPA, I have DIS and ISTA, so I'm not very familiar with the bar graphs. Certainly though, the idle air and maf values are excessive, and engine load is crazy.

Are you using the MAFand throttle body which came with the M52, or the ones from the M50? Besides the manifold, are there any other parts which are not stock? Are you using the M52's injectors? The M52's computer? Is the computer chipped?

[dworthy]

Yes they did as the Euro cars tend to come out sooner with the newer stuff, like the OBD-II was standard by MY 96 on the E-36 328i/is in the States. Here in Germany/Europe BMW brought it out sooner in YM 95. I think that BMW of North America wanted to wait until the E-39 replaced the E-34 before changing the engine over to the M-52. I get to see the newer stuff on the road here before it arrives in the States, which can be interesting as some models never get shipped there like the 2er Active touring.

[MrSkarZ]

I checked for intake vacuum leak, spray WD40 all around the intake gaskets but found nothing, removed intake airbox, throttle body NTR so far.
The mods are M50 mani that I installed 4 years ago, previous owner changed the exhaust to a M3 euro (exhaust mani, cat, mufler) and I believe replaced the 2 o2 sensors (siemens 5wk90171)
and there is a KNN air filter in the normal BMW filter box.
Injectors are the normal pink top, I never touched the DME but I suspect it was chipped in the past as the engine rev to 7000-7200 rpm.

I also replaced all lifters 2 years ago, maybe my vanos timing is bad ? Could it be that the DME was chipped for e85 ?
I just found older inpa screenshots from 2013 and fuel trims were already negatives and load around 100, maf around 15kg/h

I also have the GT1 dis tiss software but I am not used to it, what should I looking for ?

[MrSkarZ]

update: I have been trying to run DISv44 but I could not found my e36 model, everything before 1998 was missing and I got error 200.159 trying automatic detection, furthermore it is so slow compare to INPA.

So I disconnected the MAF and watch over INPA, the short term fuel trim were getting even lower than usual -67 both banks but air flow went down to 13.5 kg/h
if MAF was faulty the short term should have stabilize around 0 right ?

Then I disconnected also O2 sensors and to my surprise the Air flow bump to 15.25 and fluctuate around this value without MAF. I believe the value is coming from DME relative to idle rpm (750-800) It is pretty close to my normal MAF reading so I think that my MAF is not shot

I also checked online for new siemens 02 sensors 5wk90171 but couldn't find them, now I believe the brand is call VDO which means my sensors where not brand new 5 years ago when previous owner mount the M3 exhaust, probably second hand like the exhaust...
Is it normal for failling O2 sensors to report rich mixture ? I think it is going to be my next purchase, which brand is best NKG, Denso, or VDO ? Before that I am also going to check fuel pressure again, I have seen a video where they take fuel pressure at the pump, directly under rear sit, should be way easier than from below the car, I am expecting 3.5 bar or 50psi.

[dworthy]

TThe problem with disconnecting the MAF or oxygen sensors: The DME will go to a pre-set Fuel Map and no changes will be made. That is why you will see good number with them unplugged, a little rich, but runnable. Make sure the oxygen sensors are around .4V to 1V when the car is hot, otherwise this can point to a bad sensor. Also I would take a look at the temp sensor, as this can be sending erroneous data to the DME and putting the car at a different spot on the fuel map. Yes the FPR is rated at 3.5 Bar IIRC as I don't have the book in front of me. If you want OEM oxygen sensors, Bosch or Siemens is the best bet for those.

[MarcoZandrini]

OP, spraying WD40 on the various rubber and plastic intake parts isn't you g to find the vacuum leaks. Yes, you have vacuum leaks as noted by one of our two resident gurus, dworthy! And, our other resident guru, bmwdirtracer, has chimed in also. I'm no tech but I know when you have vacuum leaks the only sure way to find them us to have the intake smoke tested.

[MrSkarZ]

@Marco Thank for the comment, but I leave in a country where smoke test are very uncommon but I would do it if I could, the best I can do for now is spray WD40 and remove all the intake part and check manually for wear and tear. They are no good garage where I leave, most are crooks, even bmw one are shit, last time they got my car they didn't know where to plug the vanos, they connected it to IAT socket So I am very thankful for all your help here ! Maybe I can pressure test with vacuum cleaner in reverse plug into the air box or break booster line.

@dworthy Thank for advices I am going to mesure voltage at the O2 instead of from inside INPA, actually without MAF the fuel trim are even worth -67 negative. The fuel trim only got near 0 if I disconnect MAF + O2 senso
Is the temperature sensors the first one bellow cyl 1, under intake manifold ?

[dworthy]

Yes, under the #1 intake is where the dual temp sensor lives. This sensor also provides the dash gauge with info as well. So if you want to pressure the intakes, use a shop vacuum clean on blow mode, and spray around some soapy water. Where you see it bubblling up, is where the leak or leaks will be. It will be hard to get under the intake, but in that case you may just see the air comming out if it is large enough.

[MrSkarZ]

Today I had time to check the fuel pressure, I asked on another thread for normal reading at idle because I was not sure with Bentley manual and was confused from other threads, apparently idle fuel pressure should be around 3 bar(43psi) (3.5bar(51psi) minus idle vacuum from the intake 0.5bar(8 psi)) and I am reading 3.5 bar at idle and 4 bar at wot. I also check vacuum in the manifold and have been reading a steady 20 inch of mercury which also means that i have no vacuum leaks if i am correct ? I will try to plug directly the FPR to the intake manifold to see if that change something, currently with the M50 kit it goes to a T and one part goes to the intake manifold the other to muffler flap system (I will probably remove this part as I don't need it anyway)

[bmwdirtracer]

No, your fuel pressure regulator is 3.5 bar; so your fuel pressure readings are perfect. There's no subtracting manifold vacuum, or anything else. 3.5 bar under constant throttle, increasing an indeterminate amount when you jump on the throttle.

Um, I have a good online friend in your Country; smoke tests are not uncommon, you just need a better BMW shop. Generic mechanics are not going to be helpful, in your country or mine. And WD40 is pretty much useless for anything, by the way, especially finding intake leaks.

MrSkarz -- how many wires do your primary O2 sensors have?

Darin....you said that he had wide band sensors, and then told him he should have .4 to 1 volt. Help me out , my friend, because I think this is the issue. I believe we have a narrow band sensor, which reads .1 to .9 volts, but the DME thinks it's saying 4 to 5 volts !

Hang in there, MrSkarZ; Darin knows euro cars very well, and he's a truly great BMW pro.

[dworthy]

Well with the vacuum gauge reading at 20 inches, that would not indicate a vacuum leak. Still something is giving the DME erroneous information about air flow. Clean and inspect the AFM and ISV, then I would also install the new Oxygen sensors. As for that extra vacuum connector, just plug it up for now and see if that changes your reading.

Chris, I don't think the Euro models are the same as the U.S. versions as in the DME has a different fuel maps due to the fact that BMW didn't need to be so tight on their exhaust standards as the U.S. I have been known to be wrong in the past, so I am shooting from the hip.

[MrSkarZ]

Thank you for the help, there are 4 wires on my O2 sensors, I wanted to measure voltage with voltmeter but I dismount M50 again this morning. Cables are white black yellow red and inside DME it is reported as 0 to 5V range. About FPR i am very confused Bently and you said it is perfectly fine to read 3.5 bar but I find conflicting information
what about this video https://www.youtube.com/watch?v=zRrKEzPKpKs and from http://forum.e46fanatics.com/showthread.php?t=929501

Additional note/clairfication for the M52 (323/328 models). These models utilize a manifold vacuum controlled fuel pressure regulator unlike the M54 (320/325/330 models). Bentley did not properly or clearly outline how to test the fuel pressure on the M52 (323/328 models). For the M52 (323/328 models) the 3.5 +/- 0.2 bar or 50.76 +/- 2.9 PSI fuel pressure is with the vacuum line to the FuelPressure Regulatordisconnected. If the fuel pressure is tested with the engine running AND the fuel pressure regulatorvacuum hose connected, the fuel pressure should be approximately 2.8 bar or 40 PSI> Please make note of this clarification. Recently we have seen a few M52 (323/328 models) that for some reason do not have proper intake manifold vacuum to the fuel pressure regulatorcausing the full fuel pressure of 3.5 +/- 0.2 bar or 50.76 +/- 2.9 PSI to be present at idle. This WILL cause the vehicles to to run Rich and have high Negative Fuel Trims. Please pay attention to this if you have a M52 (323/328 model).

I also made this thread yesterday http://www.bimmerforums.com/forum/sh...essure-at-idle from information gathered on this forum. Maybe it is only related to E36, I am so confused...

[dworthy]

Well they are close, as the M54 motors use an integrated FPR on the fuel filter, however they are located in similar places on the U.S. models. Make sure you have proper vacuum at the FPR as I think the early 328i still have the FPR on the fuel rail. I think for the euro cars, they were never moved down to the where the fuel filter is.

[bmwdirtracer]

From Alldata:
Conversion Calculator
1996 BMW 328i (E36) L6-2793cc 2.8L DOHC (M52)
Vehicle » Engine, Cooling and Exhaust » Engine » Fuel Pressure » Testing and Inspection » Component Tests and General Diagnostics

Component Tests and General Diagnostics

Checking fuel pump delivery pressure (M52, M52TU)

Turn off ignition.



CAUTION!
The fuel in the fuel lines is under pressure (approx. 3 bar)!

Catch and dispose of escaping fuel.



Prepare special tool kit 13 5 220 (consisting of 13 5 221 and 13 5 222).
Connect special tool kit 13 5 220 to pressure sensor of DIS Tester.



Remove dust cap (1) from measuring valve on fuel rail.



Unscrew check valve (1) on special tool so that valve remains closed in fuel rail.
Attach special tool kit 13 5 220 to fuel rail and tighten down firmly so that sealing ring in special tool is sealed.



Start engine.
Screw in check valve (1) on special tool until a pressure reading is displayed at DIS Tester.

CAUTION!
Do not under any circumstances screw in the check valve up to the mechanical stop. This could damage the measuring valve in the fuel rail.



Measuring fuel pressure:

Select measuring option.
Multimeter function appears.
Select pressure test.
Read off pressure.

Test pressure,
refer to Technical Data.
Injectors And Lines, Fuel Pressure Regulator M52

86268887.gif
Switch off engine.
-----------
END
___________

Click on the little bar above "Switch off engine". I did not copy all the images, like pictures of hooking up the apparatus. But these instructions are direct copies of BMW (TIS) instructions, right down to using the DIS tester to measure the pressure.

This is also how I've ALWAYS measured fuel pressure; with the vacuum line hooked up and engine at idle, the pressure should meet the spec - and as vacuum is removed, by adding throttle, pressure should go up.

Certainly, II understand your confusion, since you're being told 2 different things. Since you have an issue, I guess you should order a new fuel pressure regulator. According to realoem, it is a 3.5 bar regulator. Personally, I think you'll end up with the same pressure, but given the information you're hearing, you'll always be uncertain until you've tried......Hey, maybe we'll all learn something new. (By the way, with engine off, pressure drops to ~3 bar; hence the initial warning of "about 3 bar". The spec for a running engine is at the end, in that image you need to click on)

[MrSkarZ]

Thank you very much guys about all this information and documentation, I really appreciate your help.
PCV and MAF were cleaned with the same results.

Yes my FPR it is a 3.5 bar, I can see it written on the side of the regulator, it is located on the end of the fuel rail near the firewall and is connected to the intake manifold with a vacuum line.
I think that DIS/TIS is right and wrong at the same time.

I am thinking about a FPR just like your finger on a water garden hose, with water flowing the more you close the hole the more the pressure.
Because the FPR is a 3.5 bar regulator it means that we want 3.5 bar of effective pressure for the optimum working condition of the injectors.
At idle there is vacuum in the intake (~8psi) therefore this vacuum help to pull the fuel from the injectors and the pressure should decrease accordingly. This is working in this particular condition because the small vacuum line connected to the FPR is pulling on the FPR spring/diaphragm, therefore increasing the passage of fuel and releasing pressure in the fuel rail. This explains why some people report 43 psi at idle for a 3.5bar(51psi) FPR.

The first wrong fact with my car is that I see 51psi pressure at idle with vacuum line connected, the second fact which is wrong with my car is that I can see 4bar at wot. However at wot there is 0 vacuum therefore the vacuum line connected to FPR release the sping/diaphragm which is now closing the passage of fuel, increasing the fuel pressure to it normal specification of 3.5 bar. My car is NA (never getting boost in intake) so I should not see higher fuel pressure than the FPR specification.

I my case the problem is probably a clogged/bad FPR or a clogged/pinched return line, that would explain a working FPR changing pressure at wot but higher pressure than spec.
And I guess the DIS/TIS is just missing the step where we should disconnect the FPR vacuum line connected to the intake manifold to see 3.5 bar fuel pressure at idle.

I am waiting for the weather to get better to work on the car, then I will report my finding.

[dworthy]

The other thing that can drop your intake vacuum is a clogged exhaust system. What is the reading of vacuum at 2k, 3k, 4k, 5k? If you see the needle bouncing around after say 4k, then it would get the exhaust looked at next.

[bmwdirtracer]

Your theory of the FPR is exactly backwards. The 51 psi reading is exactly what you're supposed to have at idle and steady, high vacuum conditions. The pressure is SUPPOSED to rise when the vacuum drops --that's when you need more fuel delivered, because the car is accelerating. Intake vacuum at idle should be ~18 inches of mercury, not 8....and vacuum is not measured in psi, it is measured in inches of mercury or inches of water. The pressure in the fuel rail does not decrease because of the vacuum in the intake runners.

The first RIGHT with your car is that you see 51 psi with the vacuum line connected, at idle, and the second RIGHT with your car is that this increases by .5 bar when you are at WOT.

DIS is correct, and you have it backwards.

[MIKYZZ4]

Can I suggest checking the- throttle valve switch -located on the throttle body, which if defective would cause the high load readings? Maybe coupled with an exhaust leak causing erroneous co2 sensor readings? If nothing found, maybe erroneous fuel maps due to damaged DME? Just my .02c

[kaivonp]

I am thinking about a FPR just like your finger on a water garden hose, with water flowing the more you close the hole the more the pressure.

I know Chris has a ton of experience, but I don’t understand how MrSkarZ’s description of FPR function is backwards. The unused fuel has to apply enough force against the diaphragm to overcome the spring pressure and exit the rail. That builds pressure in the rail, just like putting your finger over a garden hose. Manifold vacuum pulls on the diaphragm from the other side, reducing the amount of force the fuel has to apply on the diaphragm before the return passage opens—which lowers the rail pressure, just as MrSkarZ described. Is there an inaccuracy here?

So if the pressure regulator is rated to 3.5bar, to me that means that it takes 3.5bar of pressure to compress the spring and vent the unused fuel (otherwise what’s the meaning of calling it a 3.5bar regulator?). Which means that any assist on that spring reduces the pressure required on the fuel side of the diaphragm. Since manifold vacuum helps to compress the spring, doesn’t it stand to reason that rail pressure would be below 3.5bar with vacuum?

The TIS is pretty definitive, but the testing procedure Chris posted isn’t simply reading from a pressure gauge. You’re initiating a test with a computer. Couldn’t the DIS be modulating the fuel pump, injectors, or ICV to create conditions in which it expects to read 3.5bar? And that if you just used a pressure gauge at the rail schrader valve, ~3bar is what you’d expect at idle?