Defect of an injector. Or not?

Patient: E60, 530i.

This time – very unpleasant defect, where the leaking of the injector was only part of the story. But – let’s start exactly with this, because at the beginning it looked, that this is the cause of all troubles.


Piezo injectors can have several defects.

For example, leaking in closed position – this defect has a max impact in idle. Good news – this problem can be seen in adaptation data of idle of injectors ../F5/Shift+F6/F5.

Note: of course, the data of adaptations are available, if the engine works in Stratified charge and performs tests of injectors.

The injector can get stuck in open position, then the error messages regarding HPFP, misfires of cylinders can be possible, the guilty cylinder can be switched off. In the worst scenario of this defect, the leaking injector can enrich the fuel mixture of the bank in such way, that in other cylinders (which injectors work correctly) the fuel mixture gets lean and – misfires start in these cylinders. Yes, such a situation is unpleasant, but – incorrect Lambda during a problem is detected, usually, the leaking of the injector is identified (mentioned above).


But, what would you say about the following situation:

a) in any mode (in idle, in low/medium/high load, cold or work temperature) the engine suddenly shivers strongly – switches off for a moment;

b) shivering can be one or several (one after another); they can repeat after several seconds or then minutes – no regularity;

c) during shivering, the DME performs full hot-restart, in addition, this restart is with losing the indication of the tachometer, sometimes – tachometer continues to display the RPM of the engine during the restart;

d) sometimes the shivering is very strong, the engine switches off;

e) sometimes the symbol of EML lights up on KOMBI, but after a moment (5 .. 10 seconds) it turns off;

f) the error message of DME contains NO error messages, even in Info list – no error messages regarding the hot restart of the DME;

g) if shivering continues stronger, unrelated error messages are recorded, for example: regarding discrepancy of air mass and fuel amount, the error messages in DSC/DTC module (appropriate symbols light up in KOMBI), CAN error messages regarding discrepancy of the engine mode (Info list);

h) if the error message regarding discrepancy of air mass and injected/required fuel amount (DME has incorrectly processed the shivering of the engine); the engine starts to perform in very rough emergency mode.


Here, the error message regarding discrepancy of air mass and fuel:

Predicted fuel proportion is around 150 mg/stk, real: 0.

0 mg/stk fuel amount is possible in two cases:

a) the engine has stopped (or its crankshaft, camshaft sensors think so);

b) the measured air mass is 0 (the engine has really stopped unexpectedly, or the air mass meter is damaged).

Unfortunately, the attributes of the error message don’t give any clarity. The engine has shivered, several times really switched off – who knows, in which moment this error message was recorded.


In short, how the DME performs, if such an error (2D60) message is indicated. In such a situation, MSD80 concludes, that it cannot trust nor air mass meter, nor the throttle:

a) the management of the throttle is turned off. The throttle opens for around 5 degrees (what is defined by the spring – it is intended exactly for this – emergency mode), the car doesn’t react to the throttle at all;

b) DME injects the fuel amount, appropriate to the default opening of the throttle position;

c) as soon the RPM of the engine reaches more than 1300, the fuel injection is stopped, DME waits, till the RPM will drop till 1000, then restores fuel supply.

It’s very hard to drive a car, which works in such mode. RPM has to be maintained very low (below 1300) when switching to a higher gear. As soon the engine reaches 1300 RPM, the car starts to shiver. In addition – the car doesn’t react to the throttle. In the 6-th gear, the car moves with speed 50km/h and shivers regularly.

But let’s go back to our problem.


Live data review indicated the following picture:

a) during restart, DME switches off EKP, accordingly – the LPFP pressure drops; also HFPF pressure drops swiftly (it’s logical – the HPFP bypass valve has been opened);

b) during restart, DME switches off also the alternator, the onboard voltage drops;

c) directly after the restart, the engine starts to warm up the wideband probes (with PWM close to 100 %), after a moment – also the control probes, sequence – confirms complete hot-restart;

d) after a restart, the engine continues to work in Stratified charge (if it has worked in Stratified charge before), as if it has no technical trouble;

e) in the current case – directly after restarting the Lambda of 2-nd bank for a second or two is lower as for 1-st bank (0.85 against 0.95) – one of the few suspicious symptoms; after second or two the Lambda gets back to normal (1.00) in both banks;

f) during a second or two, all DME live data parameters get back to a range of correct indications – there even are no “suspects”;

g) HFPF works with pressure 150 .. 200.000 hPa, it means, in normal mode;

h) mechanical efficiency of cylinders ../F5/F7 is correct both in Homogeneous and Stratified charge mode.


At this moment I would like to mention the typical reaction of the MSD80 in case if any injector stucks in open position:

a) if the injector stays in open position: by exact PWM (which corresponds the predicted fuel consumption of HFPF), the pressure of HP contour drops swiftly – by this criteria, DME detect the HP system crash. Obviously, the HPFP is critical, the injector is damaged or the system is not airtight anymore;

b) DME performs hot-restart; turns off HPFP: rail pressure swiftly drops to LPFP pressure; in next step, DME turns on HFPF and controls the increase of HPFP pressure;

c) if the injector has stuck in open position, the HPFP pressure don’t grow till necessary value (in exact time), DME confirms the damage HP contour, by group misfires DME indicates the damaged cylinder;

d) DME disconnects the injector of the damaged cylinder, turns off HPFP, the system continues to work with the pressure of 5 Bar (emergency mode);

e) DME records the error messages regarding restart (in Info list), regarding HPFP (in Info or Error list).


Unfortunately, in the worst possible scenario, the injectors tend to “behave” in the following way – they stuck in open position, but correctly “catch up” at the moment, when HPFP pressure drops. Accordingly – when the injector stacks, DME detected the swift drop of HPFP pressure; it starts the restoring procedure, but after hot-restart detects, that everything is OK with HP contour. Coming to the conclusion, that everything is OK with HP, DME concludes restart procedure has been groundless and deletes all status bits of emergency mode and the error messages (it deletes this “unpleasant” experience from the memory). Yes, also here everything is not perfect – these restart procedures are full of software mistakes, so sometimes the error messages regarding actually correctly working components are recorded (as in the example, mentioned above – the error message regarding discrepancy of air mass and fuel amount or errors of DTC system).


The good news is the following:

a) such defects of injectors are not very popular. Most often – either the injector is leaking (which has a significant impact on idle), or it stays in open position – then DME (most often) correctly indicates the damaged cylinder. Even if the DME makes a mistake in identifying the damaged cylinder, Lambda problems are very visible (it is possible to indicate the bank, which has problems), the cylinder tests can be performed – the situation is more pleasant;

b) even in worst cases, there always are some indirect indications to other problems of the damaged injector. For example, the injector already has stuck in open position for a moment, and/or it has spilling/leaking problems.


The situation in the exact case:

The possible guilty one is the injector of cylinder No.6. This injector has stuck in open position once; at least two or three times it was leaking so intensively, that fell out of the allowed corridor (-3 mg/stk); all the time it had increased leaking in a closed position (applied correction: -0.7 .. -0.9 mg/stk).


Taking a look at the performance of the injector, I had to conclude, that the correction, applied to it, suddenly has reached -1.19 mg/stk.


After a little moment (during two-three minutes) the required correction reached -1.29 mg/stk, then already 1.36 mg/stk.


After several minutes – correction dropped to -1.02 mg/stk. Conclusion – the leaking of the injector swiftly increases/drops. One more reason to consider it as very suspicious.

After replacing the injector, all problems disappeared after a little while – the injector of cylinder No.6 was the true guilty one.


Part 2. After 24 hours

No, the car continues to shiver. Yes, DME doesn’t restart anymore. But shivering is still present. As if the engine suddenly was short of fuel, as if somebody would turn out the ignition. For a part of the second, for a very, very short moment. The engine shivers, but recovers very quickly.

For several times, the error message regarding discrepancy of air mass and fuel amount, EML in KOMBI lights up. But as soon as the engine recovers, everything is perfect!

For example, the adaptations of the injectors:

What else could have happened?

To make sure, new air mass meter was purchased to replace it immediately, if there would be some suspicion regarding the loss of its signal.


The clarity was introduced at the moment when evenly driving with speed 55 .. 60 km/h, manual gearbox: 5-th gear, the RPM in KOMBI disappeared for several seconds, after – EML lighted up and the following error message was recorded:

Such a scenario repeated for two more times, after-during several motor hours – for not a time. After these two events – for 6 .. 7 times the engine started to work in previously mentioned total emergency mode, but DME couldn’t identify the reason of crash for any of these times.


As we see in the attributes of the error message – by speed 57 km/h, RPM of the engine were 0. Which is not true – the car continued to roll by inertia.

Yes, this time the crankshaft sensor was the guilty one. For a very short moment, it stopped to send the information. For such a short moment, that the DEM self-diagnostics was not able to identify this problem This defect caused the shivering of the engine. “But the injector – what about that?” you would ask?


The leaking injector provoked the second (d)efect – restarts of DME. At the moment, when for a second the RPM of the engine were lost (as a result of the defect of crankshaft sensor), DME detected the drop of the HP counter pressure (the fuel consumption should be 0, but the leaking injector was “consuming” the fuel), and concluded, the system was not airtight anymore – and started the required recovery procedure with hot-restart, as described above. In these conditions (0 RPM) it was enough with relatively harmless leaking of the injector for DME to think, that the system of HP is not airtight anymore.

A really unpleasant coincidence of defects.