TDI Wave Acceleration: Causes and Solutions (Articles)

If the TDI engine behaves as if you are repeatedly releasing and pressing the accelerator pedal even when you're flooring it, you will likely find only fluctuating values for boost pressure, air mass, and fuel injection quantity when reading the diagnostic data, making it difficult to pinpoint the actual cause.

Most likely, these will be normal pressure fluctuations caused by other processes that enable or amplify them.

Something like this can happen, for example, when the soot reduction system is active during driving (instead of driver preference or torque limitation, see https://community.dieselschrauber.org/en/viewtopic.php?t=8670).).

Then, any change in the airflow will result in a change in the amount of fuel injected.
The intake air flow is, in turn, highly dependent on the boost pressure. Therefore, any change in boost pressure while the soot reduction system is active will cause a corresponding change in the injection quantity, resulting in a (more or less noticeable) change in engine power.

Control intervention: reduction of boost pressure -> lower injection quantity (= acceleration decreases) -> reduced turbocharger drive reinforces the initial control intervention -> the boost pressure can drop significantly below the target value, which triggers a control correction to increase the boost pressure.
Control intervention: Increase in boost pressure -> larger injection quantity (= acceleration increases) -> more turbocharger drive reinforces the initial control intervention -> the boost pressure can increase significantly above the target value, which causes a control correction to reduce the boost pressure -> see again above at "control intervention: decrease in boost pressure," etc., etc.

That completes the cycle, and we have a self-reinforcing oscillation of boost pressure and fuel injection volume – something that could potentially happen in any TDI engine.

In engines that are functioning correctly, potential "torque surge" is generally prevented by interrupting the feedback loop between air mass and fuel injection – specifically, a lower, fixed torque limit or driver demand dictates the maximum fuel injection amount, independent of (moderate) boost pressure control mechanisms.
In this way, the EDC can quickly and smoothly adjust the boost pressure without being affected by feedback effects.


Possible causes for the camel jockey effect include (roughly in order of frequency):
a) faulty mass airflow sensor
b) Extreme pressure fluctuations, for example, caused by sticking solenoid valves or severe jamming in the charging mechanism, and strong control corrections can cause the soot limit to be activated during pressure dips.
c) Blocked intake and/or exhaust passages result in low mass airflow (MAF) values.
d) The boost pressure is regulated at too low a level (e.g., due to a faulty boost pressure sensor).
e) Ineffective cooling system due to, for example, a blocked air intake; under unfavorable conditions (e.g., high temperature and altitude), this results in low mass flow rates.
f) Chip tuning that ensures the soot reduction system is constantly active, even under full load.

In case a), with aging LMMs (Linear Measurement Modules), the measured value usually decreases first in the uppermost measurement range.
Then, initially, the characteristic oscillations can only develop in the higher speed range.
As the "camelback" effect becomes more pronounced with increasing mileage and extends towards mid-range RPMs, the likely cause is an aging (or becoming dirty) mass airflow sensor (MAF).

However, not all TDIs exhibit this rocking motion when the particulate filter regeneration is active, because the feedback loop can also be mitigated by designing the charge pressure control parameters of the EDC with a certain tolerance to small target-actual deviations.
These regulatory errors are no longer (or very slowly) corrected, to the point where the overall dynamics of changes in boost pressure and fuel injection volume are insufficient to prevent a self-amplifying oscillation from occurring.


These representations are only a conceptual model that, according to the author's knowledge, corresponds to practical experience.
Corrections and additions are explicitly welcome!