Leistungsmangel und ultrahoher Verbrauch beim AEL-Motor | Posts 32+

Hello,

@ Andi from Austria.
Okay, so all the sensors/transducers have been replaced, except for the crankshaft position sensor.
The crankshaft position sensor is working correctly and provides optimal waveforms according to the oscilloscope. Also, the "angle" between the engine and the toothed wheel for the crankshaft position sensor is correct. It cannot be mounted at an angle because the camshafts are attached to the flywheel, which can only be mounted in one position (due to the screw hole pattern). While the position of the flywheel relative to the crankshaft is fixed, the angle between the crankshaft and the camshaft is not definitively lockable and must be adjusted using a feeler gauge and an alignment tool. A small error tolerance can occur when tightening the camshaft sprocket bolt.

The only camshaft/timing disc that might still be misaligned is the one inside the fuel injection pump. However, this can only be determined and adjusted by completely disassembling the pump, and it's not possible to do with simple tools.

The adjustment range of part N 108 functions perfectly mechanically (from the earliest to the latest setting) if, for example, you disconnect the connector and monitor the values using VAG-Com.
The injection timing (represented by the MWB 4 value, which should be 4) is always set late during operation, almost always around 70-97%, regardless of how much gas is being used or whether the vehicle is being fully accelerated.


I quickly recorded some data logs today regarding fuel injection timing and turbo boost pressure; anyone who's interested can take a look.

Test 1: Accelerate from approximately 70 km/h in 4th gear until the engine reaches its maximum RPM.
Test 2: Accelerate from approximately 80 km/h in 5th gear to the redline.
Test 01, 4th gear, multiple pull tests.
The last one was recorded while idling, with some occasional bursts of acceleration.

Best,
krissi

Well, the log showing the start of the injection looks "unfortunately" quite good. So, my theory is probably wrong...

Well, Ulf, that means there are already two of us who are "at a loss" (or "puzzled").



Best, Krissi.

Kristian wrote:

well, ulf, then we are two "confused" people already...

But I still think it's a matter of timing.
Has the NW setting already been checked?

If you come across a car with the same engine that runs "properly," you could perform the "basic ignition timing test" on both vehicles to rule out a misaligned crankshaft position sensor.
If the change in the engine noise when switching between early and late timing, despite the same adaptation value (-> check!), is significantly different, then it's possible that your sensor wheel is twisted, or your OT (Over Timing) sensor is misaligned or improperly mounted.

@ Kristian

As you described above (N108 timing ratio), your control unit 'wants' to set a correspondingly late injection start, which means it doesn't know exactly when the injection will actually occur (for whatever reason...).

Are the 5 'protrusions' on the flywheel all the same size, and does the control unit use the NHG signal as a reference for cylinder 4's top dead center (TDC), with the TDC for the other cylinders being derived from that, or are there markings on the flywheel, such as 'TDC cylinder X'?

Greetings,
Andi.

Hello,

I recently performed several logging sessions using VAG-Com (observed a value of 4 for measuring block 4; N108 valve for fuel injection timing).

"For me, this value is constantly in the late ignition range (from 53% to 85%), regardless of whether it's fully accelerated or slowly ramped up, or whether it's started from a low or high RPM. Also, the engine temperature doesn't seem to have any effect on it (I disconnected the radiator with a clamp => temperature up to about 115 degrees; and I also removed the thermostat and bypassed the electric fan => the temperature barely exceeded 50 degrees)."


If anyone else has the AEL configured as a 6G switch, information about the values in the main control unit (MWB 4) would be helpful (specifically, the behavior regarding engine speed and acceleration).

A friendly professor of automotive engineering at the University of Wolfsburg calculated the fuel consumption for me based on values such as fuel injection quantity, mass airflow sensor readings, injector size, number of cylinders, temperature, etc.
Result: 8.8 to 10.2 liters for multiple value pairs.

Regarding the cause of the discrepancy compared to the actual consumption (11.8 to 12.5 liters), he refrained from elaborating extensively due to time constraints.



@ulf

I used a dial indicator to check the position of the crankshaft (KW) relative to the camshaft (NW) and found a theoretical deviation of 0.09 degrees (based on the diameter of the flywheel and the "Top Dead Center" marking).

Unfortunately, due to the need for a comparative vehicle, I don't have a suitable vehicle available that is in good working order.


@ andi


The engine control unit already knows when to inject fuel, but why and where it's getting the "incorrect" values remains a mystery to me.

In any case, the usual adjustment to "early" timing during acceleration under full load, which is intended to exceed the maximum torque, is not present or measurable.
However, the characteristic curve of the VP 37 should do that.

Right now, I'm doing a "fuel system cleaning" using a chemical additive, but I don't think it will make much of a difference.

"I also noticed that the engine exhibits a significant amount of instability within the speed range of 1200 to 1400 RPM (when idling, very slowly increasing the RPM). This is noticeable through slight vibrations and an uneven, sputtering engine noise."
Starting around the year 1400, it suddenly disappears, and of course, it was also not present before the year 1200.

The engine now has 4400 km on it (4 weeks), so the initial break-in period is probably over.

With my "old" model, this behavior never occurred; it starts up gently and very quietly.


Hi Krissi,


P.S. I am grateful for any further suggestions.

Hi,

Does this part have a needle lift compensator that you could temporarily disconnect for testing?

Best regards, Rainer.

Hello,

Yes, he did, and the result was reduced performance with the same fuel consumption!

Already tried and tested.
The needle lifter is the "oldest" component on the engine, with approximately 70,000 km (about 14 months) of use, excluding the alternator and starter!
The remaining nozzles have all been fitted with new nozzle inserts.


Best,
Krissi

Kristian wrote:

The adjustment range of N 108 functions perfectly mechanically (from early to late adjustment) if, for example, you disconnect the connector and monitor the values using VAG-Com. Only during driving does the timing of the valve for fuel injection start (MWB 4 value 4) consistently operate late (almost always around 70-97%, regardless of how much gas is given or whether it is fully accelerated)

.
That shouldn't be happening!

Replace the N108 with a resistor that will not trigger an error (e.g., 100 ohms), and leave the N108 de-energized.

I'm almost willing to bet that thing will suddenly start performing properly again and consuming normal amounts of fuel. However, it will probably be quite noisy because the injection timing will always be set to the earliest possible setting.
If this works out, we would then only need to figure out why the injection process always starts late when the vehicle is in motion...

Hello!

@ Christian

You are talking about a keystroke ratio of 53% to 85%. What is the actual injection start time reported by the system, in degrees of rotation/kilowatts?

Best regards,
Andi.

Andi aus dem Ösiland wrote:

What is the actual injection start time, as reported by the system, in degrees/kilowatts?

There's already a log download available for that, and it looks fine.

I suspect that the OT (onset of torque) measured by the EDC (Electronic Diesel Control) occurs later than the actual OT, and therefore the system is generally calibrated to initiate injection too late.

It might be exotic, but in my opinion, it's the only remaining approach.

Oops! While reviewing the logs, I couldn't find the values for the injection start. I'm used to my AAT (in my C4 100) displaying the value not converted to degrees/kW (mine is 0-255, 8-bit).

@ulf

100% Agree!

Hello,

I can try the "hardware" method for N108 and let you know the results.

"However, I suspect that the OT (onset of torque) measured by the EDC is later than the actual OT, and therefore the system is generally calibrated to initiate injection too late."

I would rule that out, because if it were simply a matter of a "type" of shift moving backwards, the EDC would have to show some kind of adjustment towards an earlier reading in certain driving conditions, such as when accelerating strongly (e.g., a sudden drop in data from 70% to 40% or 50% would have to occur, which isn't what's happening; instead, the data changes very gradually).

I can't even explain this "phenomenon" within the framework of the EDC regulations themselves.

Best,
krissi

Hello.

The whole topic is really very interesting. When I see how long you've been dealing with this problem
This isn't necessarily an offer, but I currently live in GF and also drive an AEL (although it's also chipped). When I bought the car, I also had some problems with it and at the time, I read out all the control module data and compared it to Audi's specifications. One thing is certain: in my car, ALL control module data was within the specified range, except for the boost pressure and fuel injection volume (which makes sense, given the chip).

'By the way, what kind of control unit are you currently using? There are different software versions for the AEL system, so it's possible that your car is supposed to have a different version installed.'

Best regards, Mario.

Hello Gottlike,

I was working at WOB at the time, for a vehicle brand that I won't mention here. !!


The currently installed MSG (Modular Scalable Gateway) is the original one that was installed when the vehicle was purchased, and it was working perfectly fine back then.
and even two steps faster .

The current data version is D02 (which is the most common).

Control unit part number: 4A0 907 401 P.

Component or version: 2.5L R5/2VT EDC HS D02

I've tried some other things with D 12, but they all have the same effects.

However, after consulting with some tuners and a company that specializes in engine control unit diagnostics, I can say that the data version only includes changes to the exhaust emissions and starting behavior (glow plug control).
Everyone said that there were no directly noticeable effects on performance or fuel consumption, and that these effects were usually barely detectable through measurements.

It is usually only visible through a change in the characteristic curve.


Best,
krissi

Hello!

The more leisurely movement might also be due to the fact that the speed of the ESP changes, and therefore the N108 needs to be 'timed' differently for the earlier/later injection start, because there is more vacuum in the pump (or is the pressure in the ESP kept constant?).

Best regards,
Andi.