Physics behind carve ability setting?
strange last edited by strange
Does anybody have an idea how "carve ability" setting (custom shape) actually works in terms of physics? I don't understand how any motorized one-wheeled thing can allow or prevent turns. And so far I really feel no difference whether it is -5 or +5.
skyman88 last edited by
How many miles do you have under your belt? The settings are night and day.
I don't have a good answer for you on the "how" but it's been asked before. My guess is rate at which the motor and sensors are outputting and reacting. So for a tight trucks setting the board has more rapid rate of adjustment relative to inputs vs a loose setting which you can more easily turn.
This post is deleted!
@strange, I am a retired Naval aviator with a physics degree and 200+ miles on a OW+. I am fascinated by Future Motion's technology, and appreciate how hard it was to engineer and mass-produce something as refined as this product. I have played with the custom shaping quite a bit. Here is my best guess as to how the OW maneuvers.
The board has an onboard, solid-state, 3-axis inertial gyro and accelerometer platform linked to a stability computer system. The computer inputs are pitch, roll, and yaw angles and angular rates and accelerations; and linear accelerations in the longitudinal, lateral, and vertical axis. I would guess that the computer calculates miles traveled purely by counting wheel revolutions rather than by integrating longitudinal acceleration. The only physical control output the computer can make is to command the motor to apply varying torque to accelerate the wheel in either the forward or aft direction.
The only parameter the wheel and stability computer can directly affect is the pitch angle of the board by using wheel torque to balance or imbalance the forces of the rider and terrain. The computer can't do anything directly to roll or yaw the board, or to inhibit it from rolling or yawing. However, the computer can accelerate or decelerate the board forward or aft in response to any combination of fore-aft pitch and heel-toe roll. This algorithm must be fairly complex, and is the heart of making the board perform intuitively and smoothly.
I believe turning a Onewheel is accomplished by the tire being flexible, and a heel or toe roll compressing the inboard side of the tire to a smaller radius than the outboard side, and by shifting the tire contact point from directly below the middle of the board to the inboard side, both causing a natural turn. This distortion of the tire would naturally cause the board to decelerate, so it is likely that the basic stability control algorithm (neutral Custom Shaping Carve Ability setting) adds some additional torque proportional to the angle of roll to maintain speed through turns. From this neutral speed-maintaining setting, the Carve Ability scale allows adjusting to add or subtract some torque during turning. A positive setting for more torque would make the board feel stiffer, as if it was fighting the turn by trying to accelerate out of it. A negative setting for less torque would make the board more tippy, as if it was trying to help you lean into the turn by slowing.
I think Aggressiveness controls both the top speed limit (pushback onset), and the gain for how much "gas" or "brake" to apply per degree of board pitch. A lower aggressiveness setting will give a smoother, floaty ride with a lower top speed. A higher aggressiveness setting will fight harder to keep the board level and will accelerate more responsively.
The surging motor effect seems to me to be a flaw in the control algorithm. This is because it is very prominent when riding in some combinations of Custom Shaping settings, and rare in others. It does seem to be tied to accelerating or climbing, but not necessarily when the motor is working near its limits. I have experimented till I have found a relatively aggressive combination of settings that seem to work the motor hard without much surging.
I am 230 lb and have had several nosedives and many dips just short of nosedives. My impression in each one was that the motor hit its current limits and cut out for an instant to protect itself. I have read others' comments that say this is not the case, that the motor keeps working, but just hits 100% and can't stay under the feet. In one case, I know the motor cut out, because I nosedived at slow speed, was able to stay on long enough to overcorrect and rock back the other way to a full heel drag, then fell off backwards, with the board kicked out by my feet and rolling forward a good distance with no weight on the footpad, as if it was a regular unpowered skateboard with a freewheeling wheel. I had to cycle the power on the board after this event to get it to work again. This happened when I was practicing slow, tight turns and got down to about 20% battery. I have never nosedived at maximum speed, but usually on a hill climb and/or hitting a bump that was too much for the motor. I don't ignore pushback, and am very gentle and stable whenever the board starts to surge during acceleration or on an incline.
The Custom Shaping settings that I have settled on now are
Carve Ability: +5
stinkyface last edited by stinkyface
Short version of @Telomere 's thorough response. It uses yaw and roll data and more importantly the rate of change of the yaw and roll, to know if you are carving or just rolling along straight. The controller changes the rate of wheel acceleration based on the chosen carve setting and possibly also where you are in the carve cycle. Very smart!
An accelerating wheel will tend to straighten direction, so slower acceleration mid or early in the carve will allow deeper carves.
Its a good thing. This is why the board needs to know Yaw and Roll, a simple algorithm that just looks at pitch to speed up and slow down, the yaw and roll data has no purpose.
Then throw in the goofy setting and it suggests FM's algorithm varies the acceleration rate depending if you are turning front side or back side.
@stinkyface You make a good point with the idea that the motor torque output could also vary depending on how far you are into a carve cycle--perhaps slowing initially to help set the edge, and then powering up as the roll stabilizes to push the board around the turn. That would mean the computer is counting degrees of turn, as well as rate and acceleration. I do think the pitch and roll algorithms couple up sometimes, especially at the higher aggressiveness settings, and induces a persistent wobbling while going straight ahead that is not the fault of weak ankles or bumpy roads.
skyman88 last edited by
@Telomere Any time I experience that wobble it's due to my feet not being well positioned. Feels like the board is fighting due to the off balanced stance.
@skyman88, I've adopted a consistent stance of both feet close to the fender, pretty close to perpendicular to the board for better heel-toe force for tight turning, and splitting my time normal and goofy about 50-50. Wish the OW+ was a couple inches shorter, as I don't need the extra front and rear platform room, and the long projection reduces terrain clearance on hills. Wobble is rare, but if I get it, it seems to be early in the ride. Wonder if Pint does the same thing.