DME. Problems with HPFP

This time – an interesting case in one of Europe’s service centers.
I got a message – the N43 engine works unsatisfactorily. Most problems there are with a cold engine or almost cold engine. Unstable idle, shivering – the main issues.
Live data indicated massive changes (their instability) in the injector opening in idle. Checking of the Rail pressure data confirmed: Rail pressure is jumping vigorously. Jumping in a range of several tenths of bars (during one/two seconds) – such fluctuations of the pressure are not acceptable! The fuel mixture menu indicated more or less correct average fuel mixture, which means – the Rail pressure sensor works correctly. Also, low-pressure pressure – correct and stable. All cylinders are working more or less correctly; obviously: no injector is leaking, and the problem is connected with the HPFP itself.
The service specialists replaced the HPFP, but… The next cold start showed that that problem has not disappeared!
Obviously, the problem is more exotic – the connection of the HPFP should be checked: supply voltage, management signal. Control of the supply voltage with a multimeter indicated the correct voltage of +12V. Then the part of the oscilloscope came. The signal of the pump pressure regulation valve was checked using this tool.

I received such an image:

Have to admit – quite a strange signal!
Here, the connection of the HPFP pressure regulation valve:

Via FUSE F03 20A, the supply voltage +12V is delivered during pin 1 of the valve, but by pin 2: PWM signal (active=low) from the DME.
Oscillogram looks strange: during ON, the signal (on pin 2 of the valve) is not dropping from +12V to 0V, but precisely the opposite: raises from 0V to 12V:

Some ON periods are marked with red.
Obviously, the engineer has connected the oscilloscope not in the correct way:
a. GND of the oscilloscope to the body of the car;
b. probe to the valve signal;
a. GND of the oscilloscope to the output signal of the DME;
b. oscilloscope probe to the +12V.
This is an incorrect connection because it increases the capacitive load of the DME output; it also can create the short-circuit (if the GND of the oscilloscope will touch the body of the car).
The situation was actually described – the engineer confirmed that the oscilloscope is connected directly to the valve outlets. So, the true form in the output of the DME is as following:

And here – the image with the reference voltage values added:

At this moment, I don’t have any claims regarding ON cycles, but the question arises – what are these signal pikes directly after ON periods?

In the scheme below – an example of the PWM output:

Of course, the R1; Q1 model and the D1 diode are chosen according to the necessary current and voltage.
The most important this time: the D1 diode, seen in the scheme, performs two essential tasks:
a. prevents the overvoltage peaks, seen in the image above;
b. on the load (in this case – in the pressure management valve of HPFP), uses the energy accumulated during the ON period, to the mechanical work (pressing the piston – managing Rail pressure), not in heat, trying to damage management transistor.
Without the management of this diode, the transistor works in abnormal power and temperature conditions; the HPFP valve is managed incorrectly!
At this time, DME absolutely certainly has problems – obviously, the diode of the PWM output is damaged! It is due to the service specialist – replace the damaged diode or whole DME.

In the end – some nuances. Within this problem framework, the service replaced both HPFP and DME (for some time – to test it). They claimed that when even replacing DME, there were no significant improvements. Why so?
After replacing the DME, the re-adaptation of the engine is a “must-have” task! If you will install a DME with the other vehicle’s adaptation data, the result is unpredictable! The situation is even worse if any of the cars – “donor” or the “receiver” – has a leaking injector. Then misfires, shivering, vibration, turned off cylinders will be an unavoidable result. So I argue – in any case, the exact cause of the problem should be identified, only after it’s done – you can start the “curing”. The replacement of the parts by guessing will not give any result!