Select a technical image to view it in detail.
This case involves an N43 engine. The NOx system had been restored and the DME was running in full functionality mode. With the engine warm, vibration could be felt at idle. The correction of the cylinder 1 injector had reached its threshold, and the DME had stored all the error messages for these events.
Initial data and fault analysis

The engine was running in Mode 3, Stratified charge. This is logical, because error messages 3070 to 3075 are stored for exactly this operating mode. I marked the most interesting parameter: the fuel mixture is shown as “lean”. That is not unusual, because the mixture in Stratified charge mode should be lean.
The problem shows up at idle. In Stratified charge at idle, the DME tries to even out the mechanical efficiency of the cylinders by lengthening or shortening the injection pulse.
Note: in this operating mode there are two injection pulses. The first contains the adaptive compensation for the injector’s reaction time. The second contains the adaptive compensation for injector leakage (dripping). To even out the mechanical efficiency (short-term correction), the DME corrects the second pulse.
Evidently the DME had lengthened the second pulse to the maximum allowed threshold (+3.0 mg/stk) and still could not even out the mechanical efficiency of the cylinder (bring it to the minimum). It would be more accurate to describe this fault as reduced cylinder efficiency, but I have to work with the error descriptions that exist, and I cannot change them.
What can cause such a problem?
- The injector. I admit this right away: it is a rather unpopular defect.
- A mechanical problem in the cylinder. One example is a defective exhaust valve. Part of the exhaust gas escapes through it, especially at low rpm, and the cylinder’s output drops significantly.
To clarify the situation, I asked the car owner to send full diagnostic data. These are the most significant images:

This is Stratified charge at idle. The injection time of cylinder 1 is increased, especially compared with cylinder 4, the other cylinder of the same bank. It still lies within the allowed range. For myself, I have defined +/-0.50 mg/stk as an acceptable injector correction.
We have to keep in mind that the defect is intermittent. The data on the screen show the last time the engine ran in Stratified charge, while the error message is stored at the worst moment of the defect.

This is Stratified charge at partial load. Here the efficiency of cylinder 1 is reduced.
Note: in these menus the cylinders are displayed in the order 1/4/3/2.

This is idle in Homogeneous mode. The mechanical efficiency of cylinder 1 is significantly reduced, and the average fuel mixture is slightly rich.

At the same time, the injection pulses of cylinder 1 are significantly longer than those of the other cylinders.

Here the fuel quantity for bank 1 is increased both at idle and at partial load.
There is not the slightest doubt that the problem is in cylinder 1, whose efficiency is significantly reduced in all modes.
Why the situation looked suspicious
If the cause were very poor atomization by the injector, which is a fairly common injector defect, the problem should be more pronounced in Stratified charge. In Homogeneous mode the fuel has more time to mix with the air, even with poor atomization, so the situation should be much better there. Yet in Homogeneous mode the mechanical efficiency of cylinder 1 is also strongly reduced, despite the longer injection times.
Three further points made the case even more suspicious:
- The electrical parameters of the injector (charge/discharge energy and the voltage reached during injection) are correct. This confirms that the piezo elements of the injectors are healthy (../F5/Shift+F6/F1, pages 2 to 4).
- No data are available from the chemical efficiency tests of the cylinders (../F5/Shift+F6/F3). The DME evidently deleted the old test data when it detected the cylinder performance problem, and the new tests cannot be confirmed because of the fundamental performance problem of the engine.
- In Homogeneous mode the average fuel mixture is slightly rich, not lean. If the injector were leaking or delivering too little fuel, I would expect a lean mixture.
Swapping the injectors
To avoid unnecessary expense, I asked the owner to swap the injectors of cylinders 1 and 2 this time. Cylinder 2 belongs to bank 2, so if the injector is the problem, the defect should move to another cylinder and another bank.
The owner swapped the injectors of cylinders 1 and 2. To save time and effort, he did not register the injectors or re-adapt the engine. He only ran it for 5 minutes at idle and drove a short distance.
Here is the mechanical efficiency of the cylinders at idle in Homogeneous mode:

The reduced mechanical efficiency is now detected for cylinder 2.
Note: in this menu the cylinders are displayed in firing order, 1/3/4/2.

The injection times of cylinder 1 have normalized and are identical to those of the other cylinder in the same bank.
Note: in this menu the cylinders are displayed in the order 1/4/3/2.

In this menu the fuel mixture picture has reversed. At both idle and partial load, the fuel quantity is now increased for bank 2.
Conclusion
The car owner got good news: the engine itself is fine, and the problem is the injector. Such a defect is rare, even very rare. I would say it occurs in perhaps one case out of 50 to 100. Evidently the injector opens only partially, so the fuel quantity it delivers is strongly reduced.