Stuttering in 1Z: Troubleshooting Guide.

Hello!

'A while ago, I posted about the jerking issue with the 1Z engine. This was followed by several more posts describing the exact same problem. I followed all the instructions, even disconnecting the speed sensor, which ultimately worked. According to Bosch service, this is a way to test a faulty fuel injection pump. So, I disconnected the speed sensor, and sure enough, the jerking stopped (the speedometer and odometer, of course, were no longer working). My questions are: What exactly is wrong with the pump? Do I absolutely need a new pump, or can the damage be repaired with less expense? And what happens if I continue driving without the speed sensor (besides the speedometer and odometer failing)? Could it damage the engine?'

I would be very grateful for your help.

Tom.

Hi.

If the pump were seriously broken, it probably couldn't be "repaired" by simply unplugging and plugging in a speedometer sensor connector (!!).

In my case, the pump is controlled to a reduced idle speed when decelerating, which means the engine control unit is responding to the speed signal.
I suspect there might be a minor issue with the interaction between the engine control unit and the pump, so I'm going to try a DIY fix.

Okay, I'm ready. Please provide the German text you want me to translate.

Hello,

The problem with a faulty speed sensor is that your comfort functions (e.g., engine shut-off, torque limitation during acceleration) will be disabled. However, since these are also safety functions, driving without a working speed sensor will eventually lead to further damage (e.g., to the transmission or drive shaft).


@Ulf
The pump then encounters a "serious" problem: The flow control mechanism is no longer able to keep up with the rapid commands from the control unit because the control hysteresis is no longer functioning correctly. This results in overshooting or undershooting in the pump's control behavior, which manifests as jerky operation.
The speed signal is required by the control unit (ECU) for regulation (see above).
However, I have to agree with you that all other possibilities should be explored before considering a pump replacement.


I'm going to say it now, but the speed sensor could definitely be the problem. I had a friend a few days ago who came to me with similar symptoms. In addition, he had a tachometer oscillation (and it was happening on the electronic tachometer!). He then replaced the sensor, and the jerking and tachometer oscillation disappeared (lucky him!).

Bertil wrote:

The pump then encounters a "serious" problem: the flow control mechanism is no longer able to keep up with the rapid commands from the control unit (STG) because the control hysteresis is no longer functioning correctly. This results in overshooting or undershooting in the pump's control behavior, which manifests as jerking or stuttering. The speed signal is required by the control unit (ECU) for regulation (see above). However, I have to agree with you that we should try all other possibilities before resorting to a pump replacement. I'm going to claim that the speed sensor could also be defective. I had a friend a few days ago who came to me with similar symptoms. In addition, he had a tachometer oscillation (and that was on the E-tacho!!). He then replaced the sensor, and the jerking and tachometer oscillation disappeared (lucky him!!)

Hi Bertil.

For me, hysteresis means a somewhat "sluggish" reaction of the moving system to incoming commands.
Or, it can be a kind of stabilizing tipping effect that automatically pulls a system from its central position towards its extreme positions, and which must first be overcome by an external influence in order to change its state.
(I hope my "free" definition isn't completely off the mark or incomprehensible...)

Based on my experience, I can say that the tendency for sudden jolts can be clearly influenced simply by relocating the pump control unit.
A reduction in the number of elements typically either eliminates or exacerbates stuttering, and vice versa.

I find it hard to believe that the (mechanical) hysteresis of the pump can be altered simply by moving the flow control valve on the distributor piston.
I am more inclined to believe that the issue stems from a misconfiguration of the individual control parameters of the electronics and the pump, which, when combined, cause the jerking motion due to an "incorrect" signal setting in the interlocking system.

If you mean that the incorrect hysteresis is affecting the overall behavior of the system (i.e., both the electronic and mechanical aspects), then I agree with you.

A flickering digital speedometer driven by a stepper motor obviously indicates problems with speed sensing (or signal processing), which can certainly also lead to jerky movements, definitely not caused by the pump.

Hi Ulf,

... Hysteresis in an inductive system...
Ola, it's been a long time... I really need to check this out now.

Okay, so, hysteresis refers to the process of charging an inductor compared to discharging it. This results in a double S-curve that has the same starting point and the same endpoint.

In our case, the upward adjustment is achieved using a magnet, while the downward adjustment is achieved using a spring in the switchgear. I mean the hysteresis of the entire system.
You cannot influence the hysteresis by simply adjusting the quantity setting device.
However, if the engine is misfiring due to "over-fueling," then shifting is the correct way to address the issue.
If the jerking stops when the tachograph sensor is disconnected, it's unlikely that the problem is due to "overfueling."


Phew, that was a tough one. Hopefully, we haven't overloaded poor TOM with too much information.

Bertil wrote:

If the engine is misfiring due to "over-fueling," then shifting is the correct way to proceed. If the jerking stops when disconnecting the tachograph, it is unlikely that there is a "rich" fuel mixture.

Hi Bertil.

Those seem to be the symptoms.
How else would you explain this behavior (using hysteresis)?

Hi Ulf,

If the hysteresis setting is incorrect, it can cause oscillations in your pump control system, which will result in jerky operation!
This means the pump is regulating, and the control unit (STG) detects "too much (or too little)," and this process eventually escalates into a jerky motion. The inertia of the flow control valve doesn't match the specified value in the control unit, causing the control unit to overshoot the target during the regulation process, and then immediately readjust.

Such a process could be monitored using software , but this is not provided in the VAG control unit . It would then be possible to record this error as a fault code within the control unit.

I believe that both phenomena (over-oiling and incorrect hysteresis) can potentially cause jolting.

However, the oil level should be measurable using VAG COM (VAG 155*). As far as I recall from the 1Z repair manual, there's a section that specifies the target oil level and the actual oil level under certain operating conditions.

Bertil wrote:

The oil overfilling should be measurable using VAG COM (VAG 155*). As far as I recall from the 1Z repair manual, there is a section that specifies the target amount and the actual amount for certain operating conditions.

Hi Bertil,

The latter may well be true.

However, I don't think it's possible to read out an excessively high setpoint value using the 1551 protocol, because the electronics don't receive a message about the actual flow rate. The RWG feedback only includes the deflection of the control valve.

If the entire signaling system, including the RWG (reservoir water gauge), is shifted on the distributor piston such that 5 mg/KH more than intended are pumped, the RWG will still report, for example, "10 mg," while the actual amount is 15 mg.
And the 1551... can only output what the engine control unit "feels" based on the RWG signal. Therefore, an over-fueling condition caused by a throttle position adjustment should, in principle, not be measurable through diagnostics.

However, things get critical in the idle range, where the engine control unit is programmed to inject amounts of around 5 mg(?), and it would then likely "wonder" why the engine is maintaining the idle speed with what appears to be 0 mg of fuel injection.
It is possible that the jerky behavior is due to some inappropriate control strategies that the engine computer is implementing in response to such implausibilities .

ulf wrote:

Hi Bertil, The latter may well be true. However, I don't think it's possible to read out an excessively high setpoint value using the 1551 protocol, because the electronics don't receive a message about the actual flow rate. The RWG feedback only includes the deflection of the control valve. If the entire control system, including the RWG (reservoir water gauge), is shifted on the distributor valve such that 5 mg/KH more than intended is pumped, the RWG will still report, for example, "10 mg," while the actual amount being pumped is 15 mg. And the 1551... can only output what the engine control unit "feels" based on the RWG signal. Therefore, an over-fueling condition caused by the throttle position should, in principle, not be measurable via diagnostics.

That makes sense... I hadn't thought of that!

Quote:

However, things get critical in the idle range, where the engine control unit is programmed to inject amounts of around 5 mg(?), and it would then likely "wonder" why the engine is maintaining the idle speed with what appears to be 0 mg of fuel injection.

I meant the monitoring function that is probably missing in the control unit.

Quote:

It is possible that the jerky behavior is due to some inappropriate control strategies that the engine computer is implementing in light of such implausibilities

.

That's exactly what I meant with my post about hysteresis and tuning the control system (STG). The STG expects a specific control behavior from the pump, but due to a defect (or wear), this behavior is no longer as specified in the STG.
Anders kann ich mir sonst den Effekt bei abgeklemmten Tachogeber nicht erklären. Mit abgelkemmten Tachogeber sind diese Funktionen (Ruckeldämpfung, Schubabschaltung...) seitens des STG ausgeschaltet (lt. Bosch-service).
I even suspect that the vibration damping (torque damping) is responsible for this kind of vibration.

ulf wrote:

However, I don't think it's possible to read out an excessively large valve setting using the 1551 protocol, because the electronics don't receive a message about the actual flow rate. The RWG feedback only includes the deflection of the valve .

Hello Ulf and Bertil,

IMHO, it should already be possible to read out a large switch setting via diagnostics.
To maintain an idle speed of 900 RPM, a specific flow rate is required.
If the pump delivers x mg/stroke more than reported by the control unit, the control unit will still attempt to reach the target speed of 900 rpm. In other words, it will reduce the fuel flow until the engine runs at 900 rpm.
At the same time, the reported flow rate decreases through the standard flow regulator.
An overly rich fuel mixture setting should, in my opinion, be identifiable by the fact that the amount of fuel injected, as reported by the engine control unit, is significantly lower than the normal value at idle speed (in my engine, it's approximately 2.2mg per stroke, but don't take that as gospel, as I didn't completely remove the control unit).
What do you think of my theory?

Best regards, Rainer.

Rainer K. wrote:

... An overly rich fuel injection setting should therefore, in my opinion, be noticeable if the amount of fuel injected, as reported by the control unit at idle speed, is significantly lower than the normal value (in my engine, it's about 2.2mg per stroke, but don't take that as gospel, as I didn't completely disassemble the box). What do you think about my theory? ...

Hi Rainer,

Ulf also hinted at the same thing a few lines below. Let's gather the idle flow rates. Perhaps then we can get an indication of whether the mixture is "too rich" or "too lean."

Hello,

"Good idea, maybe we can then find a correlation between the fuel injection amount at idle speed (with lights and climate control off, engine at operating temperature) and a tendency for engine misfires."
"However, providing values for my own vehicle is not so straightforward, as my box adapter is quite well integrated and waterproof within the pump's cable harness. To obtain a reliable measurement, I would have to disassemble it."

For all non-OBD drivers (i.e., series/chip), using VAG-Com/1881, the values from measurement block 1 of the engine control unit would be interesting when the engine is at operating temperature, and with the lights and climate control turned off.

I was just reviewing the PDF document for industrial engine commissioning, and it states that the target values for fuel injection volume at idle are 3-9 mg/stroke?!
"With my insulin pen, I measure between 1.x and 3.x mg per injection, depending on the needle adapter... maybe I should try a measurement without the needle adapter sometime."

Best regards, Rainer.

Rainer K. wrote:

In my humble opinion, it should already be possible to read out a critical interlocking system setting via diagnostics. To maintain an idle speed of 900 RPM, a specific flow rate is required. If the pump delivers x mg/stroke more than reported by the control unit, the control unit will still attempt to reach the target speed of 900 rpm. In other words, it will reduce the fuel flow until the engine runs at 900 rpm. At the same time, the reported flow rate decreases through the standard flow regulator. An overly rich fuel mixture setting should, in my opinion, be identifiable when the fuel quantity reported by the control unit at idle speed is significantly lower than the normal value

.

Hi Rainer,

Basically, you are completely right, but in practice, things are likely to be (slightly) different, because the actual amount of fuel injected to maintain the 900 rpm will vary significantly from engine to engine - even with the same engine code.

I believe the causes could include factors such as the initial condition/wear and tear/compression, the precision of the engine control (an old, stretched timing belt leading to misaligned timing), the actual quality of the current fuel (variations in calorific value), the condition/age of the fuel injectors, even the condition of the battery (instantaneous current demand), the type of oil used, and the engine temperature, among other things. And not least, the "limited" manufacturing precision of the pump from the factory.

Overall, the spread of the displayed values is likely to be so large that it will be difficult to definitively conclude that a signal box position is too high in any individual case, unless "gigantic" errors occur, such as a displayed value of only 30% of the usual value according to the repair manual.

Hello everyone!

I have only been reading these very informative pages so far, and I have already found many helpful tips for troubleshooting.

Right now, I'm stuck with this persistent vibration issue between 1300 RPM and 2000 RPM on my Passat with the 66kW TDI engine (1Z), which has 185,000 km on it.

Okay, I'll post the measurement data block 1, as measured with VAG-COM.

Idle speed.

903 revolutions per minute.

Injection amount.

2.6 mg/H

Voltage sensor, control slide position.

1.36 V

Coolant.

88.2°C


I will be installing a new fuel injection pump in the next few days, and I hope that will finally solve the problem.

Best regards, Detlef.

Hi Detlef,

I'm eagerly awaiting the result, like a cat on a hot tin roof.

Quote:

Injection amount: 2.6 mg/H. Voltage sensor, control slide position: 1.36 V

Hi.

Compared to an industrial-grade TDI (AFD) system, the pump might be set too rich.
Its target values at idle are 3-9 mg and 1.5-1.95 volts.

A "simple" adjustment of the quantity setting device might resolve the issue at no cost --> see the link above in this thread, or "DIY quick fix" in the TDI error database.