Audi A4 TDI: Jerking at Low RPM - Fix!

Hello everyone, I'd like to welcome you all to the forum. This is my first post.

Okay, I have the following problem;
'For some time now, my A4 1.9L TDI 96KW has been experiencing occasional short jolts during driving at low RPM (1300-1900 RPM), regardless of speed or gear. This happens when I gently press the accelerator, usually when driving at a constant speed, but also when accelerating slightly. The jolts are brief, as if the engine is momentarily hesitating, and there are no accompanying noises. Imagine quickly tapping the tip of your foot on the brake. This happens sporadically, sometimes more, sometimes less. I've already taken it to the dealership, but they couldn't find anything wrong (they also can't reproduce it, as I can't force it to happen during a test drive!). This also occurs regardless of whether the engine is warm or cold, or whether the outside temperature is warm or cold. At first, I thought that since I hadn't driven long distances recently, the engine just needed to be properly 'broken in' again. Yesterday, I was on a business trip to Munich (200km from my home), and I drove it a bit more aggressively. Even when I was driving slower, at 80 km/h on the highways, I still experienced 2-3 of these jolts throughout the journey, but they occurred at longer intervals.'
Does anyone else have similar problems or experiences?
It would be great if someone could get in touch who might know something or have a hunch!

Best regards, Claudius.

Hi Claudius (is that really your name?),

These could potentially be the first signs of issues with the Relais 109, which is known to cause problems, especially with VP-TDIs.

A stuck EGR valve could also be a possible cause.

Otherwise, I would suggest checking the error memory (or having it checked) and posting the results here – then we can see what's going on.

Hello, I assume the error memory will be clear, since the car was already at the dealer's.
Instead, I would investigate the idle speed control system, considering a faulty dual-mass flywheel, defective pressure control elements, or a defective transmission control module (TCM) as a last resort.
First, I would like to have several expressions for the idle speed control.
If the polarity (+ or -) changes for individual cylinders, then the crankshaft position sensor (CKP sensor) is likely defective.
If only one or two values deviate significantly, then it is likely that one or more PD (partial discharge) elements are defective.

You have to imagine it like when you quickly tap the brake with the tip of your foot.

Is the brake light switch definitely and absolutely ruled out as a cause of the problem?
If it's set to an extremely tight tolerance, I could definitely imagine something like that happening due to vibrations (resonance).

CU Gremlin.

Hm, sorry guys, I missed.

What is the year of manufacture of the A4, and what is the mileage? Does it have a 5-speed or 6-speed transmission?

If the 109 is faulty, the engine would stall and not sputter. Additionally, the fuel sensor would overflow, just like with a malfunctioning LL (low load) control system, etc.

The mechanic should check the mass airflow sensor (MAF) to verify the amount of air being drawn in. Then, they should check the functionality and sampling rate of the EGR valve.

If that doesn't solve the problem, the brake light switch and clutch switch located on the pedals should be checked for proper function and potential looseness, and possibly replaced.

@Julian: Not quite correct: Deviations in the LL (Low Level) control are not necessarily stored in the error memory, unless the tolerance limits are exceeded.
'In cases of defective PD (pre-catalytic converter) components or a faulty ZM (flywheel) module, this threshold is often not exceeded, which is why no error code is typically stored.'

@Julian: Not quite correct: Deviations in the LL (Low Level) control are not necessarily stored in the error memory, unless the tolerance limits are exceeded.


Okay... the engine is known to be a bit rough, which means that the idle speed control might be reaching its limits. However, I find an idle speed range of 1300-1900 RPM to be impossible with a correctly adjusted pressure regulator.

'@Julian, it's not just regulated at idle; the fuel injection quantity per cylinder is constantly being increased or decreased, depending on what the engine control unit (ECU) calculates for each cylinder between each reference point. To identify this, you have the values from the idle speed control, which show you where there are deviations.'

Oops, it seems you're mixing up two different things there.

1. Idle speed control - is active when the engine is idling and primarily serves comfort-related purposes.
2. Torque compensation system - is active during idle and is designed to compensate for differences in cylinder torque, improving both comfort and exhaust emissions.

Both operate independently and are speed-dependent. The main differences lie in the control strategy: #1 controls minimal amounts, while #2 operates at the maximum amount and maximum speed. This function requires the most processing power.

Incidentally, distributor pumps do not have a flow compensation system; the control system is simply too slow for that. It only works with low loads.

Regardless, even if the UI (user interface) were completely malfunctioning, the motor wouldn't collapse when the accelerator pedal is pressed. There are still three other motors working. The effect is described as 'like lightly tapping the brake pedal.' So, first check those; they are sometimes set very precisely.
Then I would focus on the air system (mass airflow sensor, EGR, turbocharger).

CU Gremlin.

Regarding the jolting sensation after the acceleration phase:
/viewtopic.php?t=1671&start=11

I also had that problem until the EGR system was deactivated... which I didn't want. Afterwards, the engine control unit (ECU) was replaced, and two errors were fixed at once.
Okay, I'm ready. Please provide the German text you want me to translate.

Thank you for the many responses.

@ulf
Yes, I believe I'm quite attractive; it's at least mentioned on my ID card!

The car was manufactured in April 2001, has a 5-speed transmission, and has 47,000 kilometers on the odometer.

I installed a new LMM (lambda sensor) three weeks ago, so I don't think that's the reason. I'm less likely to believe it's the LL (lean/rich) control system, because it's always just a single jerk during driving. I probably forgot to mention that!
'What really surprises me is that this sudden jerk creates a stronger braking effect than simply lifting off the accelerator. This suggests that something must be briefly locking up. I can't imagine it's the brakes themselves, or that a faulty brake light relay or some other electrical issue could cause such a brief braking effect, so I doubt that's the cause. Therefore, the braking force is more likely coming from either the engine itself or the transmission. Am I missing something here? Please correct me if I'm wrong.'

Best regards, Claudius.

Hm, if I read this, one could almost think that the brake assist system is overreacting, or?

I would completely disassemble and clean the AGR (Abgasrückführung system). I also have that planned for my car soon. A friend of mine also has a 2001 8E AWX with similar issues as yours.

Brake light switches and clutch switches are still my favorites.