Welcome to Tesla Motors Club
Discuss Tesla's Model S, Model 3, Model X, Model Y, Cybertruck, Roadster and More.
Register

Version 7.1 AP Hand on Wheel Sensing

This site may earn commission on affiliate links.

RDoc

2021 Prerefresh Model S
I started a new thread specifically about the AP's sensing of hands on the wheel since there's starting to be some FUD thrown.

Before this devolves into a total flame war, I decided to do some actual testing and now I think I know what's happening. My experiment was to drive on a 3 lane highway on AP and see if and when the Hands on Wheel alert came up.

To start with, as I've seen with this update before, the only time I got the warning was when the road markings deteriorated too much. However, some people with updated Teslas have reported the warning appearing every 3-5 minutes, so it seemed worth pursuing some more.

I always drive in AP with one hand lightly gripping the wheel so I can feel its motion and to be ready to take over if it does something crazy. This time though I tried completely hands off driving. That started the warnings at just about exactly 5 minute intervals. My conclusion is that, in addition to cases where the car isn't sure what's going on, if the car can't sense you holding the wheel for 5 minutes it emits the warning. Some people may be getting the warning if they are letting the wheel slide in their hand, but I find that uncomfortable. It seems like holding it tightly enough that you can feel the wheel moving and it moves your hand and wrist is enough. I wasn't even aware it was an issue till I started playing with it.

It is my very strong opinion that using AP without a hand on the wheel, in addition to being plain misuse, is dangerous to the occupants of the car and surrounding traffic. I understand that there is also a new check that prohibits AP if no one is in the driver's seat as well, and I'd put this check in the same category.

I guess if it's important to you to be able to drive without holding the wheel at all, or for that matter, being in the driver's seat, then this update may disappoint you.
 
Hmm, I drove for almost 15 minutes on the highway today and didn't get a nag. No hands on the wheel but nearby (on knees right near the wheel). However, I did change lanes a few times. (The car did, I mean). I'll have to check tomorrow to see if I can repeat your findings.
 
Hmm, I drove for almost 15 minutes on the highway today and didn't get a nag. No hands on the wheel but nearby (on knees right near the wheel). However, I did change lanes a few times. (The car did, I mean). I'll have to check tomorrow to see if I can repeat your findings.
That's interesting indeed. Perhaps any interaction with the AP counts as hands on wheel and restarts the clock.
75mph 25 mile stretch no lane changes no speed changes "Hold steering wheel" every ~3 minutes. Never had the nag on that stretch or interstate until 7.1 today. Beautiful weather pristinely clean car nothing different other than 7.1.
Did you have your hand(s) on the wheel?
 
I usually keep my hands off the steering wheel and definitely feel there is more nag now with 7.1.
However, at least my casual observation has been that it does not come on every 3-5 minutes.
Where I noticed it more is in curves and unfortunately there's a lot of those around here.
 
g. Just spend 30 minutes testing this.

Interstate road, 12-lanes divided (6 each direction), every 3 minutes for 12 minutes (i.e. 4 times) then I got off. Exactly at the 3minute mark.
Local road, 6 lanes (3 each direction, grass divider, though the car thought it was an undivided highway and restricted AP), every 1minute for 4 minutes (i.e. 4 times).
Then I held the wheel for about 5minutes, and still on the local road it let me drive for 2 minutes, then nagged, and then back down to 1 minute. I think it might be giving me leeway if it sees me holding the wheel.


Up until this evening I was in the "nah, there's no nag" category, but I think there is, as it's repetitive (in whole minute intervals), and I also think the nag is different on different types of roads.
 
There has been FUD about this entire thing ever since the release of V7.0 from all sides.

For whatever reason the regulators in California got the impression that there was a timed nag, but there was never a REAL timed nag in V7.0. People talked about there being one, but it was never proven. No one with a stop watch ever came to any conclusions. To me it seemed like an engineering solution. Where it used the corners to determine if someone had their hands at the steering wheel, and used that moment to tell the driver. I say this because even in shallow corners with great lines it would sometimes ask. I didn't mind it, and had no problems with it.

Now with V7.1 there does appear to be a timed nag. Where the nag gets reset if it's unable to detect a hand at the steering wheel. This detection is made difficult because the sensor requires torque to be applied, and of course some people won't apply torque even if they have a hand on the steering wheel. This is especially true on long straightaways where there isn't any corner. Overall this solution seems a lot more lawyerish than engineering because the solution isn't a good one engineering wise.

I personally haven't tested it as I don't have the update yet. But, when I get the update I intend on testing it. With V7.0 it seemed to require a bit more torque than I really thought was necessary to make the warning go away. I would say it was almost 25% of the torque required to disable Auto-steer, or perhaps more.

At to my own behavior its all situational dependent. Sometimes with lots of fast dense traffic I don't even use it, and choose to use TACC only. In more mild situations I have one hand on the steering wheel, but not fighting it. In the middle of nowhere I don't have my hands on the steering wheel. If my hands are cold I have both hands on the steering wheel with the heated steering wheel on. Yep, all that chocolate I've had over my life has led me to being a better driver. :)

Anyways in an attempt to cut through what may or may not be FUD lets just stick to the pro's and con's of this update without attacking people for how they drive. We all know there are three kinds of drivers. The crazy people, the morons, and you the perfect driver.

Pro's of 7.1 AP
1.) It supposedly tracks lines better when they sudden start to fade.
2.) It handles the exit off ramps better where it doesn't dive for them.
3.) Better positioning in lane on right lane.

Con's of 7.1 AP
1.) It now has a timed nag every 5 minutes to remind you to have your hand(s) at the steering wheel if it's unable to detect them. Applies to situations/people where either no steering torque is applied or very little.
2.) It's not always going to correctly determine what type of road you're on and the speed limit of the road. This will lead to situations where you can't use the full AP on roads you could with V7.0

Anything I'm missing?
 
Last edited:
Yes, you appear to be missing the fact that multiple people with 7.1 have used AP and not reported a timed nag. Their posts are scattered around the various 7.1 related threads, some of which are dozens of pages long. Here is a snippet from one such post:
I drove for 30 min @75mph this morning on the highway without touching a single thing. No Nag at all.
Of course multiple people have reported a timed nag in 7.1. Their posts are easier to find because they sometimes start a new thread about it.
My theory is that the nag occurs on roads where AP does not have good data on, and does not occur on roads where AP does have good data on. But that is just a theory.
 
Last edited:
My theory is that the nag occurs on roads where AP does not have good data on, and does not occur on roads where AP does have good data on. But that is just a theory.

I disagree. I have an interstate highway here, where I'm sure it has good data on, that I get a timed nag at the 3-minute mark, repeatable.

I agree though that it's based on the types of roads, but I don't think that it's based on if AP has good data or bad data. I just haven't figured out what it's based on just yet...

Interstate road, 12-lanes divided (6 each direction), every 3 minutes for 12 minutes (i.e. 4 times) then I got off. Exactly at the 3minute mark.
Local road, 6 lanes (3 each direction, grass divider, though the car thought it was an undivided highway and restricted AP), every 1minute for 4 minutes (i.e. 4 times).
Then I held the wheel for about 5minutes, and still on the local road it let me drive for 2 minutes, then nagged, and then back down to 1 minute. I think it might be giving me leeway if it sees me holding the wheel.


Up until this evening I was in the "nah, there's no nag" category, but I think there is, as it's repetitive (in whole minute intervals), and I also think the nag is different on different types of roads.
 
Con's of 7.1 AP
1.) It now has a timed nag every 5 minutes to remind you to have your hand(s) at the steering wheel if it's unable to detect them. Applies to situations/people where either no steering torque is applied or very little.
2.) It's not always going to correctly determine what type of road you're on and the speed limit of the road. This will lead to situations where you can't use the full AP on roads you could with V7.0

Anything I'm missing?
From various reports, it appears that there is a definite timed nag, however the timing appears to be variable (and we don't know the variables) some have timed it with a stopwatch at 5 minutes, others at 3, others have reported none, or 1. Normally I would conclude that it just doesn't exist, as I did on 7.0, but the timed ones are to the second, and even the people who are in the "no nag" camp seem to indicate "slightly more nag than 7.0" so I'm more inclined to believe that there is a timed nag, but that it varies depending on either speed driven, or type of road.

I'm still in the camp of not installing 7.1 as the pros are very minor compared with some pretty major cons.
 
From various reports, it appears that there is a definite timed nag, however the timing appears to be variable (and we don't know the variables) some have timed it with a stopwatch at 5 minutes, others at 3, others have reported none, or 1. Normally I would conclude that it just doesn't exist, as I did on 7.0, but the timed ones are to the second, and even the people who are in the "no nag" camp seem to indicate "slightly more nag than 7.0" so I'm more inclined to believe that there is a timed nag, but that it varies depending on either speed driven, or type of road.

I'm still in the camp of not installing 7.1 as the pros are very minor compared with some pretty major cons.

And that's why I started a thread to get people to comment on it. My current running theory is that the nag is based on the speed limit on the road. I've had both repeatable 3 minute and 1 minute nags so far

How and why to nag (AP V7.1)
 
I was in the camp of no nag until I just did some tests. My experience is similar to OP and others and it seems to be a difference of literally not touching the steering wheel versus a hand in some way touching the steering wheel. I chose a stretch of I-25 between Denver and WY where it's very straight for long stretches and the speed limit is 75. I stayed in the right lane with AP set to 79 mph throughout the test. On the way out I kept my hands and feet away from everything, didn't change lanes or adjust a thing. I didn't even touch the 17". I might as well have been asleep or dead as far as the car was concerned. Within a couple minutes I was following a pickup truck going about 65-70, so most of this was below the speed limit. Under that scenario, my stopwatch came up with a timed nag at 5 minutes three times in a row.

I turned around and did the same stretch coming back. This time I rested my left hand on my knee with my fingers around the steering wheel but relaxed completely. I could feel the very minor adjustments that AP makes even on a straight road but the wheel could slip through my fingers. About half way back I was stuck behind a truck going 60 but I didn't touch anything - because it's for science! Again I didn't interact with the car in any way (except the left hand relaxed around the wheel).There was not a single nag for the 17 minute return trip.

Based on this, I'd say that the timed nag would mostly affect those who are not touching the steering wheel at all for extended periods of time. That actually seems reasonable to me. The car probably should try to find out if you're asleep (or worse) occasionally, just in case. It also seems prudent to keep a hand in contact with the wheel so you can feel when things are about to go wrong. If you're reading a book or knitting and wait until your body feels the lateral Gs from the AP doing something it shouldn't, you're probably going to be way too late to that party.
 
Why doesn't someone at Tesla end this debate (or contribute to it) by telling us what was designed to happen? And a lot of other debates/disputes, fights.

Typical Tesla. Just like they didn't end any other debate on this forum in the past. If enough people complain they'll release a statement, but the truth is that Mr. Clown is correct. This is annoying, but you really should have your hands on the wheel, at least lightly. And if with a light touch there is no nag, then while this is restrictive, it does make sense.
 
Typical Tesla. Just like they didn't end any other debate on this forum in the past. If enough people complain they'll release a statement, but the truth is that Mr. Clown is correct. This is annoying, but you really should have your hands on the wheel, at least lightly. And if with a light touch there is no nag, then while this is restrictive, it does make sense.
Yes, I agree. A light touch with one hand is a lot better than having to hold the wheel, even with one hand draped on the top as so many people drive now. I really don't know why people are complaining.