The situation of this case:
N43-series engine; restored functionality of the NOx system, and DME works on full functionality mode. For the warm engine, the vibration can be felt idle. The correction of the injector of the 1-st cylinder has reached the threshold value, DME has recorded all error messages regarding these events.

As you see, the engine was working on Mode 3: Stratified charge – it is logical because the error messages 3070 till 3075 are recorded exactly for this performance mode. I marked the most interesting parameter – the fuel mixture was is if “lean”. How to understand this? The fuel mixture in Stratified charge mode should be lean!
As you see, the problem is identified in idle. In Stratified charge/idle, DME tries to even out the mechanical efficiency of cylinders, increasing/reducing the injection impulse’s length.
Note: there are two such injection impulses in the current work mode. The first impulse contains the adaptive compensation for the reaction time of the injector; the second impulse contains adaptive compensation regarding the leaking (dripping) of the injector. To even out the mechanical efficiency (short-term correction), the second injection impulse is corrected.
Obviously, DME has increased the length of the second injection impulse till the max allowed threshold (+3.0mg/stk), but still could not even out (reach the minimal) mechanical efficiency of a cylinder. In this case, it would be more correct to mention in the attributes – reduced efficiency of the cylinder. But I have to work with the error message descriptions, which are available – I have no power to change them.
What can cause such a problem?
a. the injector. I have admitted immediately – quite an unpopular defect;
b. mechanical problems with the current cylinder. For example – a defect of the outlet valve: part of exhaust gases are flowing out via it, (especially) with low RPM, the cylinder’s performance is significantly reduced.
To clarify the situation, I asked the owner of the car to send the full diagnostics information. Here, most significant images:

Here, you see Stratified charge/idle – the injection time of the 1-st cylinder (especially comparing to cylinder No.4: the second cylinder of this bank) in idle is increased. Still, it also fits in the allowed range. I have defined for myself that +/-0.50mg/stk is an acceptable correction for the injectors. Here we have to understand that the defect is “fluctuating”, which means: the data, which we see on screen, show the last time when the engine was running on Stratified charge, but the error message is recorded in the worst moment of its “lifetime.”

Stratified charge/partial load. Here, in the mode of the partial load, the efficiency of the 1-st cylinder is reduced.

Note: in these menus, cylinders are displayed in the following order: 1/4/3/2

Here, the idle, Homogeneous mode. As we see, the mechanical efficiency of the 1-st cylinder is significantly reduced—the average fuel mixture – slightly rich.

Simultaneously, the injection impulses of the 1-st cylinder are significantly longer than for other cylinders.

Here, the fuel amount of the 1-st bank is increased both in idle and in partial load mode.

There is not the slightest doubt – the problem is exactly in the 1-st cylinder; in all modes, its efficiency is significantly reduced.
In this case, the situation was suspicious. If the problem’s cause were very bad atomization of the injector (quite a popular defect of the injector), the defect would be more pronounced directly in Stratified charge. In Homogeneous mode, the situation was supposed to be much better because, in this mode, the fuel has more time to mix with the air (even in case of bad fuel atomization). But, as we see – also in Homogeneous mode, the 1-st cylinder’s mechanical efficiency is strongly reduced, not taking into account long (increased) injection times.
Even more, the situation is double suspicious because:
a. the electrical parameters of the injector (charge/discharge energy, the voltage reached during injection) are correct – it confirms good “health” of the piezo elements of the injectors (../F5/Shift+F6/F1, page 2 .. 4);
b. data of the chemical efficiency tests of cylinders (../F5/Shift+F6/F3) are not available. Obviously, DME has deleted old test data when detected the problems with the performance of the cylinder, but the new test data are not confirmed due to fundamental performance problems of the engine;
c. in Homogeneous mode, the average fuel mixture is slightly rich, not lean – as it would expect if the injector leaks, not enough fuel.
To avoid unreasonable investments, this time, I asked to switch places of the injectors of cylinder No.1 and No.2. Cylinder No.2 refers to Bank 2; if the problem is the injector, the defect should move to another cylinder + another bank.
The owner of the car swapped the injectors of cylinders No.1 and No.2. To save time/energy, the injector’s registration and the re-adaptation of the engine were not performed – only 5 minutes in an idle and short driving session.
Here, the mechanical efficiency of cylinders in idle, in Homogeneous mode:

In the menu, you see – now the reduced mechanical efficiency is detected for the 2-nd cylinder!
Note: in this menu, cylinders are displayed in firing order: 1/3/4/2

Here, the injection times of cylinder No.1 have normalized – they are identical to the second cylinder of the same bank.
Note: in this menu, the cylinders are displayed in 1/4/3/2 order.

In this menu, you see that the picture of the fuel mixture has changed to the opposite. In both idle and partial load mode, the fuel mixture should be increased to the 2-nd bank.

At this time, the car owner gets good news – everything is fine with the engine itself! The problem is exactly with the injector. Yes, such a defect is rare. I would say – even very rare! Maybe even in only one case of 50 .. 100 cases. Obviously, as a result of the defect, the injector opens only partially, the fuel amount injected by it is strongly reduced.