A neighbor recently had the battery in her car die. When she got a jumpstart, the steering in the car was not normal. She was unable to accurately describe it, but it was bad enough that she was afraid to drive the car. The steering apparently alternately pulled left then right.
The repair shop that replaced the battery said that new cars have electronic steering mechanisms that require “recalibration” when the battery is disconnected.
I hadn’t heard this before and both my neighbor and I feel it is more likely that the shop took advantage of her. But the car drives well now so she is somewhat mollified.
It’s possible. It would be bad engineering if the steering ECUs didn’t keep their calibration data in nonvolatile memory, but it’s possible. That “alternately pull left then right” could be a sine wave from bad PID coefficients (too much gain), and it’s possible they used a self tuning PID or adjustable parameters.
Would be abysmally bad engineering work by the automaker, who made this car?
Other possibility I’m thinking is the surges of current from the steering assist motor wasn’t getting properly supplied by the battery. So the car would run but whenever a sudden surge of power was needed, the battery voltage was plummeting under load like failed lead acid batteries tend to do. Possibly this was leading to a feedback loop with the controller.
Unfortunately without basically this whole apparatus, the schematics and source code, the actual hardware, and so on, it’s extremely difficult to accurately diagnose a system problem like this. I have fought similar bugs and usually the root cause is never what you think it is.
The way a mechanic fixes things is usually they just guess what the problem is or look up the symptoms somewhere, then start swapping components until the problem goes away.
When I replaced the battery in my wife’s '08 Jetta, the steering assist light came on, and it took a few minutes of driving before it went out. I didn’t notice any significant change in the steering behavior, though.
With all the computers on modern cars, jump-starting is not as simple as it used to be. Get it wrong and something can get fried by a voltage surge.
Sometimes a system does have to re-learn after the battery is disconnected or goes flat and this usually just takes a few minutes. I have no idea whether this could apply to steering though.
it’s possible that if it has something like active nibble control (yes, that’s what it’s called) that it was re-learning. or was using a previously stored strategy.
Thanks all.
My first step is to see what car she is driving so that I have some useful information to pursue this.
Then I think I will try checking youtube.