I wrote about B38 software problems with idle and persistent vibration in some other entries. Now I’m adding to this pity story with strangeness, which were noticed when examining the performance of the B38 model for the Year 2022 closer. Needless to say, the vehicle has the actual integration in May 2022.
First, clearly software peculiarity – the flywheel adapts, but in poor quality. OK, I don’t say anything regarding higher RPM (above 3000). In this mode, probably, there are not enough MCU resources. Probably, the engineers do not “strain” because the flowrate differences should be predictable (measured in lower RPM) and can “go” according to the (injector’s) adaptation maps. Interesting, though, that the sensor data are stable, which means – the interruption system of this MCU is with a large reserve! But the problem is (also) in the range of low RPM. Even after as if completed flywheel adaptation, In the Overrun mode, we see the following “picture”:
Here, it can be clearly seen that the curve of the 2nd cylinder (green) is different from the curves of the two other cylinders. Checking this after a particular time (during which the adaptation data should be “brushed” if it would be necessary), the situation stayed identical.
Here I have to add that the flywheel data, “adapted” in the factory, were even less correct. To make the situation even more fun, as usual – no adaptation learnbit, no other data.
How so? Why is the data reference of all cylinders not evened out accurately? This is a question of the software and, further – hardware problems.
At the moment, I don’t have information regarding more than one new B38 (B38 of the Year 2018 also was inspected, but it had different DME, different software, and no Expert mode available); the data will be supplemented during the time. But this exact, as if correctly (by the manufacturer’s point of view) adapted, utterly new B38 gives more questions than answers. For example, Valvetronic. Here, the idle:
As we see, the position of the motor rotor (blue color) is constantly jerking. Such jerking (and even more impressive) happens in all stationary conditions, including the warming-up phase (when the Valvetronic is not in active use).
Here is the idle position of the eccentric shaft:
We see that DME changes the eccentric shaft’s position in a small range all the time. B58 engine, for example, behaves more correctly.
Why don’t I like such hysterical jerking? The main reason – is faster wear out of separate hubs. In this case – plain bearings of the Valvetronic motor and worm gear. When the worm gear and the axis and plain bearings rapidly change the turning directions and have large variable axial loads, the hubs have large radial loads instead of swiftly increasing speed from 0.
Replacement of the Valvetronic electro motor costs around EUR 400. It will be harsher if the toothed wheel of the eccentric shaft wears out.
Question – why is Valvetronic so actively taking part in the short-term regulation? Yes, it is possible that the software is to blame. But the reason could be much more straightforward – in such a way, the electric throttle is spared.
In different releases of B38, slightly different electrical throttles are used. Still, all are united with one nuance – the electric motor, which mages them, is a “regular” DC collector motor, which could be used for some electric toy, not for an essential hub of the car engine. From this aspect – yes, Valvetronic has a 3-phase non-collector motor and oiled worm gear. Accordingly, this hub can handle more “torment”.
Sorry for the sarcasm. The longevity of both hubs in the existing modes and performance gives me serious doubts. But, I understand – when can you expect a car, which costs “only” at least 25’000 EUR.