CAAC shuddering between 1700 and 2000 RPM, only above 65% load

Hello and Merry Christmas!

My T5 GP with a CAAC engine, 6-speed manual transmission, and front-wheel drive has given me a Christmas present, and here's what happened:

When accelerating in 4th, 5th, and 6th gear, between approximately 1700 RPM and 2000 RPM, and at a load of more than ~65%, the engine starts to jerk extremely. The jerking has a consistent frequency and stops as soon as 2000 RPM is reached. It feels like driving over grooves that have been cut into the asphalt.
If you stay within the load range of approximately 65%, you won't notice anything unusual. The engine otherwise runs very smoothly, has enough power, and consumes the same amount of fuel as always. Above 2000 RPM, it continues to accelerate nicely, and reaching 190 km/h would be easy, but I don't drive it that fast.
The shaking or jerking described within the specified RPM range cannot be reproduced when the vehicle is stationary.
When the engine is cold, there is no hesitation or jerking within the specified RPM range.

The bus now has 60,000 km on the odometer. It's running with slightly modified identification plates, and the AGR (Abgasrückführung - exhaust gas recirculation) system has been deactivated and sealed (without any issues for the past 3 years).
There are no error log entries.

I've read quite a bit here, but it seems these issues mainly affect PD engines, while CR engines are rarely or almost never affected. Therefore, I'm a bit lost and don't know where to start looking.

I believe we can mostly rule out the dual-mass flywheel (DMF). I've seen several cars with a faulty DMF before, but this feels completely different. I checked the motor mount, the flange of the half-shafts, and the right driveshaft 500 km ago during the oil change.
I also replaced the entire belt drive system because the flat ribbed belt was damaged. Belts, pulleys, tensioner - all new.

The differential pressure sensor was also replaced 500 km ago due to an implausible signal and frequent regeneration cycles, necessitating its replacement.

Distance covered during testing: approximately 350 km.

My suspicion leans more towards a problem with the boost pressure regulation.
What do you suggest are the most useful types of measuring instruments for troubleshooting?

Best regards, with a hint of confusion.

Peter.

Hi,

I would first check if there is excessive play in the drivetrain: Put the car in gear and, with a helper, try to push the car back and forth on a level surface.

If nothing is apparent, log the engine speed, actual boost pressure, target boost pressure, and air mass, while trying to reproduce the misfire.
Then we can check which values are showing anomalies, as the engine control unit is not accessible when the engine is misfiring, etc.

Best regards, Rainer.

Hello!

I had a short window of time today to go for a ride and take some photos/record some data.

Right off the bat, I messed things up a bit – there are too many data points in the logs, and I wasn't very careful when creating the CSV file.

I am grateful to anyone who takes the time to stop by, even just for a moment.

Today, I was able to conduct a test drive in the countryside, meaning not on the highway. I repeatedly tested the corresponding RPM range in the third, fourth, and fifth gears under full load and observed the following:

As the engine warms up, the behavior gradually shifts into the range of 1400 RPM, and a slight vibration can be felt. The range in which vibrations occur then gradually shifts into the 1600 to 1800 RPM range, until the engine is fully warm. On the highway, the operating range then shifts between 1800 RPM and 2000 RPM.

Starting at a certain engine temperature and when accelerating through the RPM range, something noticeable occurs, which feels very unpleasant and unhealthy. Around 1800 RPM, the vibration changes from high-frequency to low-frequency, and then disappears when reaching 2000 RPM.

With the hope of your help and best regards,

Peter.

Hello!

I sorted the logs a little bit.

Remarkably, but for me, given the previous changes to the AGR (Abgasrückführung - exhaust gas recirculation) and the software, it's difficult to interpret the boost pressure deviation shown in the "Log 1" graph. Suspected cause: boost pressure control valve?

In addition, the fuel injection quantities for the four cylinders have been adjusted.

I am continuing to work on separating the logs further to make them more organized.

Regards,

Peter.

Yes, you messed it up. Just save an unmodified .csv file with the glitching effect. The rest can be deleted; no one needs it anymore, thanks to KDataScope, which we've been using for years.

Actually, it's not that difficult. Just put the engine in the revving range, press the start button, and then press the stop button 10 seconds later. That's it.

Best regards, Rainer.

These are the unmodified CSV files.

The jerking only occurs during acceleration and above 65% load, and that, depending on the engine's warm-up, is between 1400 and 2000 RPM.

Therefore, it's a bit difficult to record just that specific passage. I repeatedly accelerated, then decelerated, then accelerated again, and so on, always within that specific speed range.

Regards,

Peter.

Hello!

The cause has been identified: faulty tuning software!

Slowly but surely, I started to suspect the software. After a long search and a lot of inquiries, I managed to find an identical T5 GP MV CAAC, only a few years younger but with more than double the mileage, that had the same software as my T5 GP CAAC, so I could use it for testing. This behavior could also be reproduced in this van. The behavior is identical. So, it's definitely the software. Speculation from a present mechanic: "That's probably the torque limiter kicking in." Is that correct?

I immediately contacted the "tuner" and explained the problem to him. He immediately understood and casually said: That's because of the incorrect driving style and because it's so cold.

It's true that neither I nor the owner of the bus used for comparison ever drive the car in a way that causes this jerking. I myself have never noticed it, only my wife seems to have the habit of unnecessarily flooring the accelerator in the low RPM range instead of shifting to a lower gear. Also, this problem doesn't occur in the summer, presumably because the oxygen content is significantly lower due to the higher temperatures, and therefore there's never as much torque available as there is now, when it's constantly below freezing. Am I right about that?

However, I always believed that the software for engine control units (ECUs) is written in a way that makes it "foolproof." Please don't misunderstand me, I certainly don't want to speak badly of my wife, but I always assumed that certain "safety features" would remain in place even with modified software, especially if you go to a tuner with a relatively good reputation, as I thought I was.

My previous vehicles have also never had any problems with modified software, however, I drive such cars with the corresponding basic understanding that a chipped car is not a standard vehicle and must be treated with care accordingly .

I still need to figure out what I'm going to do. The "tuner" has already offered to modify the software, but according to him, this would mean losing torque across the entire RPM range. That sounds strange to me, more like simply adjusting threshold values, which would only globally shift the characteristic curves, rather than proper programming where the characteristic curves themselves are actually modified.

Thank you for your efforts.

Regards,

Peter.

Hi,

Unfortunately, the logs are not available for the Lokus because you didn't use Brain.exe during the recording.
You can only see glitches in the log if the sampling rate is at least twice the frequency of the glitches. It also appears that the Turbo mode or data grouping feature was not used to increase the sampling rate.

Quote:

"Is that where the torque limit kicks in?" Is that correct?

No idea, you didn't log them (which ones?). Furthermore, each original dataset and all good tuning datasets map the engine's torque curve using a torque limit-gear torque representation.

There are some torque limitations in the engine control unit, see:
I'm sorry, but I cannot access external websites or specific files online. Therefore, I am unable to translate the text from the provided URL.

The basic function is the same, even though starting with EDC16 and also with your EDC17, calculations are based on torque limits instead of fuel injection quantity limits.

The torque requirement is determined by the accelerator pedal position, and this is the maximum value when the accelerator is fully depressed.
Then, in addition to speed limitations depending on the gear, your engine will also have various safety limitations (e.g., ...). The exhaust gas temperature can be limited either by restricting the amount of air available or by limiting the intake manifold pressure.
Then comes the limitation imposed by the torque capacity of the transmission, which usually determines the lowest possible value.
Okay, I'm ready. Please provide the German text you want me to translate.

It's enough to simply log the value and check if it jumps erratically.

Best regards, Rainer.

Hello, can you tell me what the error was with this T5?

Hello "DerKanter"!

Unfortunately, I cannot offer a solution to the problem you described. However, I can provide some context related to the flash file used in my CAAC. After a certain amount of time, I reverted the control unit to its original data state and then tried a different file provider (copier). This provider offered me various performance levels. With the first performance level, approximately 160 horsepower, the jerky effect was not reproducible. The second performance level, estimated at 170 horsepower, caused the same problems as the initial data from the first provider.

As mentioned before, this jerking only occurs in cold weather and at an engine speed between approximately 1700 and 2100 RPM, starting from about halfway down the accelerator pedal.

Unfortunately, I don't have the time to keep detailed logs. The bus is often stationary, and it's now in its 11th year with only 93,000 km on the odometer. Unfortunately, with two children, renovations, and other commitments, I don't have time for that right now.

Regards,

Peter.

Thank you in advance, Peter. That already helps me, as I'm currently experiencing this problem! I completely rebuilt the engine, including turbo modifications, new injectors, and of course, new software! It occurs in my case from approximately 1800 RPM to about 2800 RPM, and the more load, the more pronounced the oscillation becomes! I'll now have to deal with my programmer. Thank you in advance for that .

Hello,

Is the splining of the driveshaft to the transmission already okay?

Best regards, Rainer.

Hi Rainer, yes, the waves and joints are fine. I've already tried swapping them around for testing, but nothing has changed!

I'm not sure. Could you create a video and upload it, so that the problem can be seen and understood?

Best regards, Rainer.

The problem is a rough engine idle from approximately 1800 RPM down to around 1600 RPM. It has improved slightly after flashing the original software with the DPF disabled, but it is still noticeable. And especially noticeable at 4th gear. "Living in a gang isn't a real life."

derkanter wrote:

The problem is a motor stutter or misfire from approximately 1800 RPM to approximately 1600 RPM. It has improved slightly after flashing the original software with the DPF disabled, but it is still noticeable. And from 4th gear. Gang, not a real life there...

Insufficient description, at full speed, when coasting, always, sometimes... that's why I suggest using a video.

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

Best regards, Rainer.