R2 High Speed Instability.

[emroch]

When I went on a test drive this past week, I had a little bit of wheel shake at 65ish MPH. I chalked it up to a front wheel balancing issue. But now that I have read all of the posts in this thread, I think it is likely a steering wheel position sensor variance in a few R2s. Since it happens at UHF is on but not if it is off, it can't be a wheel balancing issue.

My theory is that there is a wheel position sensor and that there is a tiny amount of variance in how the sensor is mounted. The software likely breaks the 360 degrees of steering wheel rotation into some small arbitrary fraction of rotation so that when it needs to turn right or left, it turns the wheel X number of rotation fractions to get that amount of vehicle course correction. But if the sensor is mounted a hair left or right of dead center, UHF is constantly reading that the steering wheel rotation is not a whole fraction of the rotation it is trying to accomplish. So it thinks it has gone too far, corrects by going back one increment, which is thinks is too far because the sensor is not reading as a full increment, which then causes UHF to correct again ad infinitum. Hence a tiny shimmy back and forth at the frequency that it uses for 55-70mph steering wheel course corrections. (The frequency and size of course correction increments likely varies according to vehicle speed which it is why it is felt at some speed ranges and not others.) Could this be fixed in software? Sure, but if I'm correct, it would mean that the 55-70mph course correction increment would need to be increased which would mean a major change in how their UHF software works in order to fix a small percentage of vehicles. The best fix would be to adjust the position sensor on the few vehicles that are having the issue.

Just a theory...

If we assume this theory is correct, another software solution would be to add tolerance to the measured position value. If it is really shimmying between adjacent increments, then the increments are small enough to be within the steering column's slop as the motion does not translate to the wheels. They could just tweak the UHF algorithm to accept a "close enough" position value or set a minimum delta required before any correction is applied. It doesn't need to make 1° adjustments all the time, it can wait until the calculated adjustment is 2° (or whatever the next increment is).

Sponsored

 

[windshieldfarmer]

The fact the problem has continued beyond the original few thousand VINs makes me think this is indeed some type of undesirable mounting variance of the steering position sensor. While an OTA fix would be ideal for Rivian from a cost perspective it would seem adding too much steering variance could introduce undesirable minor ping ponging within the lanes. I guess alternatively they could manufacture replacement sensors with predetermined mounting offsets for affected vehicles. Regardless…please fix my R2. This is irritating.