A yoke or a joke?

[M0del3At5paceX]

I have ordered my plaid S with the yoke.

Having said that, here are some thoughts.

1. I have spent the last two days driving my car pretending I have a yoke, and not touching the top half of the steering wheel. I find it mildly annoying at best and dangerous at worst.
2. Driving schools will correctly teach you that 10 and 3 hand positions allows for the most precise steering. Holding the yoke this way results in only half of your hands actually on the steering wheel which is tiring. It also means placing your hands over the touch sensitive buttons on the steering wheel, potentially inadvertently pressing all sorts of buttons.
3. Driving schools will correctly teach you not to put "hand over hand" while turning the steering wheel. This is completely unavoidable with a yoke.
4. While having turn signals and some controls on the steering wheel is beneficial because it allows you to activate those functions without moving your hands, having physical buttons is preferential because it allows you to feel the button without looking down, and also decreases accidental button pushes. Touch sensitive buttons look cool, but require you to look down to confirm that your hand is actually on the button. This negates the benefit of having the buttons on the wheel in the first place (other than having no stalks). It also increases accidental pushes, and no one wants to be that guy driving down the road, unaware that his turn signal is on.
5. When the yoke is upside down during a turn, the buttons on the yoke are reversed. Up is down and left is right. While this seems obvious, it is dangerous. The left turn signal is now on the right and vice versa. Cars with gear paddles fixed to the steering wheel have the dangerous situation where the up-shift (on the right) is now the down shift which can cause the car to loose control if pressed at the wrong time and the car is downshifted during a turn causing a dangerous oversteer. In an emergency situation, muscle memory takes precedence.
6. To emphasize the point further, in an emergency situation or an impending accident, grabbing at air, where the top of the steering wheel used to be could be fatal. Half a steering wheel means twice the chance of not being able to steer out of danger.
7. Furthermore, a steering wheel with nobs on it has a good potential of getting caught on clothing or other objects when turning. This could inhibit or prevent a successful turn and cause accidents.
8. While it is true that planes and formula one cars have a yoke, it is also true that they can’t make tight 90 degree turns either. They don’t back into tight parking spots at the supermarket. A pilot never ( and cannot) do a full lock to lock 5 rotations of the steering wheel to do a three point turn for example. This potentially could be solved by a speed sensitive steering algorithm, but would have to be quite extreme to compensate.
9. The model S is expensive, and a large sedan. So the people likely to purchase it are likely to be older people who have money and have families. These are the people who are least likely to learn new tricks, or adopt gimmicks that might compromise comfort or safety.

Stop thinking about the yoke like a steering wheel. It’s not going to spin around 360 like a steering wheel. The amount of control it provides to the front wheels will be inversely proportional to speed. You won’t have to spin your wheel around to park. Heck, my model 3 is currently in my garage with an upside down steering wheel due to some last minute maneuvering. No more of that.

 

[TravelFree]

Stop thinking about the yoke like a steering wheel. It’s not going to spin around 360 like a steering wheel. The amount of control it provides to the front wheels will be inversely proportional to speed. You won’t have to spin your wheel around to park. Heck, my model 3 is currently in my garage with an upside down steering wheel due to some last minute maneuvering. No more of that.

I would believe your claim if you can show us some evidence it is true. The Cybertruck video actually shows your claim is not true.

Anyway, it doesn't matter what I think because I'm not in the market to buy another S. So, for those who want it, have fun...

 

[CapeOne]

Just because an airplane uses a yoke doesn't make it better in a car.

Assuming yoke steering does meet government regulations, I think it would be a mistake for Tesla not to also offer a traditional round steering wheel as an option.

 

[drkeys]

I have ordered my plaid S with the yoke.

Having said that, here are some thoughts.

1. I have spent the last two days driving my car pretending I have a yoke, and not touching the top half of the steering wheel. I find it mildly annoying at best and dangerous at worst.
2. Driving schools will correctly teach you that 10 and 3 hand positions allows for the most precise steering. Holding the yoke this way results in only half of your hands actually on the steering wheel which is tiring. It also means placing your hands over the touch sensitive buttons on the steering wheel, potentially inadvertently pressing all sorts of buttons.
3. Driving schools will correctly teach you not to put "hand over hand" while turning the steering wheel. This is completely unavoidable with a yoke.
4. While having turn signals and some controls on the steering wheel is beneficial because it allows you to activate those functions without moving your hands, having physical buttons is preferential because it allows you to feel the button without looking down, and also decreases accidental button pushes. Touch sensitive buttons look cool, but require you to look down to confirm that your hand is actually on the button. This negates the benefit of having the buttons on the wheel in the first place (other than having no stalks). It also increases accidental pushes, and no one wants to be that guy driving down the road, unaware that his turn signal is on.
5. When the yoke is upside down during a turn, the buttons on the yoke are reversed. Up is down and left is right. While this seems obvious, it is dangerous. The left turn signal is now on the right and vice versa. Cars with gear paddles fixed to the steering wheel have the dangerous situation where the up-shift (on the right) is now the down shift which can cause the car to loose control if pressed at the wrong time and the car is downshifted during a turn causing a dangerous oversteer. In an emergency situation, muscle memory takes precedence.
6. To emphasize the point further, in an emergency situation or an impending accident, grabbing at air, where the top of the steering wheel used to be could be fatal. Half a steering wheel means twice the chance of not being able to steer out of danger.
7. Furthermore, a steering wheel with nobs on it has a good potential of getting caught on clothing or other objects when turning. This could inhibit or prevent a successful turn and cause accidents.
8. While it is true that planes and formula one cars have a yoke, it is also true that they can’t make tight 90 degree turns either. They don’t back into tight parking spots at the supermarket. A pilot never ( and cannot) do a full lock to lock 5 rotations of the steering wheel to do a three point turn for example. This potentially could be solved by a speed sensitive steering algorithm, but would have to be quite extreme to compensate.
9. The model S is expensive, and a large sedan. So the people likely to purchase it are likely to be older people who have money and have families. These are the people who are least likely to learn new tricks, or adopt gimmicks that might compromise comfort or safety.

they actually don't teach that anymore due to airbags the 10 and 3 hold position results in loss of control for the steering wheel when air bag deploys. most of them suggests 4 and 8 now. so you can maintain steering wheel direction with air bags deployed. plus it's more comfortable with arm rests or knee as resting points for long term driving.

as long as the steering wheel is a closed loop which from the new shorten yolk steering attachment that it is just a slightly different shape it totally works with most people's driving style.

 

[ucmndd]

Stop thinking about the yoke like a steering wheel. It’s not going to spin around 360 like a steering wheel. The amount of control it provides to the front wheels will be inversely proportional to speed. You won’t have to spin your wheel around to park.

Why are you stating this as a fact despite any evidence to support the claim?

 

[M0del3At5paceX]

Why are you stating this as a fact despite any evidence to support the claim?

Because using a yoke as a wheel is dumb. Tesla is very software driven. They'd only go this route if there was something going on behind the scenes with the software. In short... it's logical.

 

[ucmndd]

Because using a yoke as a wheel is dumb. Tesla is very software driven. They'd only go this route if there was something going on behind the scenes with the software. In short... it's logical.

Oh. You can state baseless speculation as fact because Tesla is logical. Got it.

I lol’d

 

[M0del3At5paceX]

I would believe your claim if you can show us some evidence it is true. The Cybertruck video actually shows your claim is not true.

The Cybertruck that's been driving around and in promotional videos is a prototype. That means not everything you see in it or on it, or how it behaves is necessarily the same as what the production units will have.

 

[M0del3At5paceX]

Oh. You can state baseless speculation as fact because Tesla is logical. Got it.

I lol’d

No different than ALL of the speculation going on. Just pointing out that the posts that think you'll use this yoke the same way you use a typical steering wheel (like rotate it 360 degrees) are not thinking properly about it, or about Tesla. Decades ago, perhaps, I would have agreed, but in the current world of control software, heck no. It's just an opinion of people who can't look past what they have become used to.

 

[sillydriver]

I’m skeptical about the street use of yoke steering, but since my order is for a plaid+ I’m going to have the benefit of being able to look at the experience of others for at least a half a year before deciding whether to take delivery of a car with a yoke or not. And I’m particularly skeptical if it turns out you have to rotate the yoke more than 90 degrees each way. In this vein I think the worst problem isn’t the yoke by itself: it’s the combination of the yoke and crucial buttons on the yoke, especially the turn signal buttons that would be directionally reversed when the yoke is upside down, while all the other buttons would be exchanged side-to-side too. And I think it’s interesting that this remains nearly as much of a problem if the car ships with a wheel that has buttons on the spokes or on the hub. Just having an option to get a wheel isn’t going to solve the button problem.

But if we assume there is a yoke, then we have both the button problem and all the ergonomic problems relative to a wheel that are involved in cranking a yoke two turns from lock to lock. These problems are overcome if the yoke is limited to 90 degree rotation each way: 1/2 turn from lock to lock. If you do that, so that your left hand can always be on the left side of the yoke, then muscle memory of the angle from fingers to thumb would probably let you hit the correct turn signal button even though the yoke is deflected from vertical. Yes I know the Cybertruck allows multiple turns lock-to-lock, but then I believe the twin problems of crossing arms on a yoke and trying to get the buttons right becomes very severe.

But the yoke (or wheel) also needs to give you the whole usable control range of the car. At very low speed this means the 1/2 turn from lock to lock yoke must reproduce the steering motion of the existing Model S, which is 2.05 (round it to two) turns from lock to lock. That leads to one quarter the leverage and four times the steering effort, which requires four times the power steering boost relative to the human input force to overcome at low speed. Power steering had better not fail, but then it’s not good if it fails in an ordinary heavy car too.

It would be extremely dangerous if the steering maintained the same very quick ratio at high speed, so the ratio needs to be variable: so variable that I’m not sure a mechanical system could do it. The steering likely has to be by wire.

Real-world testing is needed to tell whether this works for drivers, but it may be the best way to set the variable steering ratio is to make it the function of speed that would leave the lateral (or centripetal) acceleration constant at any given yoke angle at speeds above a certain speed.

What I mean is that, for example, a 90 degree deflection of the yoke to its lock point would produce, for example, a 1g lateral acceleration in a turn at whatever higher speed the car is going. I pick the round number 1g for this example because its roughly the maximum the tires could handle (although in reality they could handle a little more). At very slow speeds, that 90 degree yoke deflection would just turn the car as tightly as its turning circle allows.

To be specific, my MS owner’s manual says the turning circle is 38.8 feet or 11.8 meters curb to curb. Measuring from the center of the car, let’s say the turning circle has roughly a 10m diameter and 5m radius. The centripetal acceleration is a = v^2 / r, which means v = Sqrt(a r). 1g acceleration is 9.8 m/s^2, so v = Sqrt(9.8 x 5) = 7 m/s = 16 mph. You would be near the limit circling on a skid pad at 16mph at maximum steering angle.

So up to 16mph, which is the speed where you get 1g lateral acceleration in as tight a turn as the car can do, the steering holds a fixed ratio where a 90 degree yoke deflection produces the minimum turning circle. Above that speed the ratio could change to keep the car turning at 1g at 90 degrees deflection. Since r = v^2 / a, and a is fixed at 1g, then r = v^2. This means that if the car’s speed doubles, the turning radius to deliver 1g quadruples. So if we double the speed from 16 to 32mph, the steering ratio is reduced, and the radius of the turn is increased, by a factor of four. As a result, at 32mph the steering ratio is the same as in the current Model S where it takes two turns from lock to lock to steer in the tightest circle from curb to curb. If we continue this scheme, then at for example 128mph the steering ratio would be only 16th (the steering would be 16 times slower) than in the current Model S.

This may be too radical a change in ratio with variation in speed, and too big a departure from normal steering, for a driver to get used to it, although it would have the biofeedback benefit of being able to dial in a desired feeling of sideways acceleration consistently at a given yoke deflection at all but the lowest speeds. And it would also make it harder to lose control by rapid movement of the yoke (or wheel) at high speeds. The best variable-ratio scheme may be some compromise between this function and fixed-ratio steering where having just a 90 deflection produce the car’s maximum steering angle would be very dangerous at high speeds. Perhaps the ratio could vary with speed from 4x the current car’s (to avoid requiring more than 90 degree yoke deflection going slowly) to 1x the current car’s ratio at high speed.

Another reason (other than the need to avoid crossed hands and upside-down buttons) to think a variable ratio may be used is that I recall (or I think I recall) that a few years back when Elon introduced the new roadster he spoke of augmenting driver steering input with a computer’s drive-by wire adjustments that would keep the car on an efficient line while driving on a track and reduce the likelihood of a loss of control. That could become part of a comprehensive drive by wire steering system with a variable ratio. There would likely be different modes for street and track, with different ratio profiles.

But my message here is that the combination of the yoke and vital function buttons on the yoke/wheel makes having more than 1/2 turn from lock to lock very problematic, meaning a decreasing ratio at higher speeds is needed, and would be needed even if the steering wheel stayed round but had the all those important buttons on the hub and spokes.

 

[beatle]

Stop thinking about the yoke like a steering wheel. It’s not going to spin around 360 like a steering wheel. The amount of control it provides to the front wheels will be inversely proportional to speed. You won’t have to spin your wheel around to park. Heck, my model 3 is currently in my garage with an upside down steering wheel due to some last minute maneuvering. No more of that.

You can have progressive / variable rate steering with a round wheel as well.

Tesla did a pretty good job with the S when it came out. They appealed to a lot more people than just those that wanted to save the environment because it was just a good car. They should be careful not to alienate the people who buy their cars. Time will tell just how many people refuse to accept their design language. I know I would not own a model 3 due to the barren interior. This change may be polarizing enough to drive others away from the purchase as well.

 

[M0del3At5paceX]

You can have progressive / variable rate steering with a round wheel as well.

Tesla did a pretty good job with the S when it came out. They appealed to a lot more people than just those that wanted to save the environment because it was just a good car. They should be careful not to alienate the people who buy their cars. Time will tell just how many people refuse to accept their design language. I know I would not own a model 3 due to the barren interior. This change may be polarizing enough to drive others away from the purchase as well.

Some may be turned away. Some might now be attracted to it because it adds to the cool factor. After a while, people will see that it works and come back.

 

[mycroftxxx]

Aerospace engineer here...

The comparisons with aircraft yokes is off base, because turning the yoke (or for stick-equipped aircraft, moving the stick side to side) controls the roll of the aircraft; it does not steer the aircraft on the ground. I won’t go into the details, but essentially all aircraft steering on the ground is done via the pedals (one for each foot) and maybe a lever or knob (the “tiller”) for low speed maneuvering. On modern fly-by-wire aircraft, the on-ground steering control is designed to provide the pilot with predictable and non-twitchy steering characteristics across the speed envelope; it may be designed to provide constant yaw rate or lateral acceleration per pound of force applied to the pedals, as an example; this requires adjusting the effective gearing as a function of ground speed.

I personally see no reason why a modern automobile steer-by-wire system can’t incorporate the same kinds of speed-adaptive control as an airplane uses, and a yoke instead of a wheel may work work well as the control inceptor for such a scheme. But if so, it won’t be because many aircraft use yokes for totally different kinds of control.

 

[beatle]

Some may be turned away. Some might now be attracted to it because it adds to the cool factor. After a while, people will see that it works and come back.

I still would not consider the model 3 until I have a binnacle. The aftermarket has now solved that problem more or less - a shame that Tesla has not. I'm sure there are 3rd party manufacturers salivating at the market potential for fixing Tesla's design mistake by offering a traditional wheel.

 

[TEG]

 

[Black/Black MS]

After having seen the Yoke I would never buy a Tesla without one. I hope it has adaptive steering so that the yoke doesn't turn more then 90 degrees, but I'd only buy that option either way.

 

[Wol747]

>>Soon, if no one stops us, those of us who have gone from 100% mechanical human control to 100% automated human control will describe how a transition from high muscle intensive mechanical wheel/yoke to sidestick demanding light, precise pressures.<<

I think you've actually put your finger on the reason sidestick control is unlikely to be used in cars!
From the many videos I see of aircraft landing in strong crosswinds I see far too many examples of overcontrol - and almost always on sidestick aircraft. As a strong righthander I can't even use a mouse with any accuracy with the "wrong" hand.
Vehicle control requires arguably more sensitive and constantly accurate control in traffic than flying an aircraft, and I question whether this yoke - looks very Star Wars - thingy will ever get put in the current generation of cars. The control requirements are completely different to an airplane. The thought of trying a three point turn with traffic backed up, a yoke and a car that "knows" whether you want forward or reverse is enough to keep me up all night!

 

[Spacep0d]

I think variable ratio steering should help and must be getting employed here, though I’ve seen no mention of it.
At slow speeds a 90 degree turn of the wheel gives full lock of the front wheels and at 90 mph a 90 degree turn maybe gives 20 degrees from center. Then, you never have to take you hands off the wheel.

Tesla hasn't mentioned it but variable ratio power steering already exists in modern vehicles, where slower speeds offer more assist and less so at-speed. I presume that this is the solution for the yoke, something I've mentioned in my own posts as well. It won't be your typical captain's wheel with the easy-turn knob. It'll be something new, different, and knowing Tesla it'll probably be better.

The one thing I'm curious about is multi-point turns and how the 'Guess-O-Matic®' transmission is going to work here. Is it going to be slow and ponderous or nice and quick? How obvious will it be when a gear is changed? I imagine that a chime will announce the gear change (which could be annoying), or hopefully something subtle but noticeable.

All I know is that I'm a yoke fan and I want one for my Model 3! My car is a road-going spaceship and a yoke would be a great addition!

 

[drtimhill]

I've been a pilot since the late 60s and every airplane save one that I've flown has used a yoke. I like yokes and don't really see a problem having them in what I drive on the ground. I like both of my current Ss so don't anticipate buying a yoke model anytime soon but do hope I get to test drive one just to see what it's like.

Try this .. put some harmless hand cream on parts of the wheel on your model S where the yoke would NOT have part of the wheel available. The try to drive your model S without touching any of the hand cream. See how far you get.

Your airplanes have totally different requirements to a car steering wheel. You might as well argue that boats can work with huge spoked wheels and so we should use those on cars too.

 

[Spacep0d]

Oh. You can state baseless speculation as fact because Tesla is logical. Got it.

I lol’d

Actually, logic is our best tool here. A yoke which only serves as a cosmetic replacement for a steering wheel is not just illogical, it's a hazard. Why would Tesla do this?

That's like making a car with tires that mimic a wagon wheel. They're both wheels but imagine the shock of people used to wagon wheels wondering how a modern wheel would work? This is a bit of an absurd comparison, but we cannot expect new tech to retain the exact same functionality in a different form. An EV is a car, but it's fundamentally different from an ICE vehicle, fundamentally better!

It's Tesla. They've been innovating all along. The entire line of cars is innovative, EV technology included.

It's going to be good, because they will test it and it likely won't make production if it's found wanting. They've likely already tested this. Doesn't the new Roadster 2.0 already have the yoke, the one where Tesla was giving people test rides?

The only way a yoke makes sense is not as a wheel dressed up like a yoke, but for the yoke to have functionality for both driving and FSD (greater driver instrumentation visibility, smaller footprint). This must necessarily include some kind of variable ratio power steering (already in modern cars) which would support the functionality of a yoke rather than working at cross-purposes with it.

If Tesla sticks with it it's probably going to be a winner.