Thrown off after nosedive with no pushback warning (nose dive, nose-dive)
kwatts last edited by
I've had the OW for a few months, love the thing, but there is definitely something going on with nosedives in extreme mode. I've had it happen and ate San Francisco Asphalt more then once, at least three times. Same deal with what has been described on this thread - extreme mode, 50+% power, usually flat, then out of no-where the front hits and I go flying.
OW devs: We should have the ability to record all the readings (accel, speed, etc) as sort of a flight recorder of data. I'm sure a lot of us would even send it to OW to analyze. There is something going on here, and if you'd just give us the tools - SDK? - we could figure out what's going on from the data.
TonyGDTLA last edited by
@fabuz I may be wrong, but after watching the video, it looks like you had a pretty heavy pushback and your tail touch and you foot came off sensor all in a second. I've had my tail touch a few times. I found it helpful to practice in a controlled area recovering from a tail touch.
chabis last edited by
@TonyGDTLA when do tailtouches occur? whenever you get a heavy pushback? or how were you able to practice them? :D
TonyGDTLA last edited by
@chabis sometimes the pushback pushes and without thinking your center of gravity could shift back. And if you lean back in too fast it the tail will touch. I would practice on grass (helmet and stuff) accelerate fast then hard break. Tail will touch and drag. But stay balance and try to recover.
chabis last edited by
@TonyGDTLA good to know... had a few tailtouches up until now but could recover from all of them. still i was worried not to be able to do it every time so this "exercise" should present an opportunity to get "used to it" as much as possible haha. dang it. why's my OW being all weird and stuff. would love to try this right now :(
ted last edited by
@kwatts Agreed. The option to enable/disable a data recorder from the app would be awesome. Or, release an SDK and see what the community builds.
There is something going on. It has happened to me several times now: Happily carving at a steady speed on asphalt at ~50% power; feet planted rock steady on both sensors; nose-dive out of nowhere. Ouch.
Problem is I love riding OW too much to not take the punishment.
callenj357 last edited by
I had a pretty bad spill this weekend. Nosedive out of nowhere. I ride often and have a good amount of experience on the board. I am very mindful of my feet and the sensors. Early on when i read about this happening to others, i thought it was related to lack of experience. This is definitely a real issue. For me, flat (very hard) surface, 50%+ on the battery, extreme mode. Luckily i was wearing a helmet and wrist guards.. so only have some bad bruises and raspberries on the left ankle, calf, knee, hip and shoulder.
I do not want to believe it was a malfunction.. because if that is the case, makes me a bit nervous to ride again. But i love it so much. So i ordered some more safety gear.. will wear impact shorts, knee and elbow padding.
Will report back if it happens again. Lets hope not. but i think we have enough reports of this occurring with riders who have had some decent time in the saddle for FM to look into it.
An experience and some questions:
Been riding for a few weeks, always in classic mode, new board so new firmware. Have ridden on smooth pavement, rough pavement, hard dirt, soft slightly muddy dirt, grass. Until this weekend never had the OW do something that I didn’t understand. When first learning I didn’t understand how to work the sensors so I was run over or thrown off in reverse a few times. I believe I now have a good handle on these and can stop and dismount easily in 1 spot without really thinking about it. I don’t ride too fast and don’t think I’ve felt pushback.
This Sunday I thought I would get in a ride before the rain started again here in SF. Streets were dry. I live in some moderate hills and cruised around and up and down not over 6-7mph. Then, with 60%+ battery remaining, I was going straight on a relatively flat street, probably a bit slower than my average speed. While it happened pretty fast, my recollection is that without any warning and without losing my balance. the nose of my board ate the pavement. Wearing helmet and wrist guards, I launched forward – as a snow sports guy I put my right (forward) arm out by reflex. The board continued to travel forward at a 45% angle to the right, thunking loudly against a garage door where it stopped. I dislocated my right arm and spent the rest of the day in the UCSF Emergency Room high on Dilaudid and Fentanyl.
Obviously I would like to understand why this happened. From reading this thread I see some theories but the root cause is unclear. Given the occurrence from time to time of similar experiences, I believe it would be prudent for FM to provide collection of and selected access to real-time actuator/sensor data. If the OW lacks sufficient memory I don’t think it would be hard to log across Bluetooth to the phone. I also think it would be in the company’s best interest in order to improve and refine the behavior of the OW. I would understand a potential reluctance based on litigation risk but it needs to be done and the lawyers would just have to do their best to protect the company, as they do in all motor vehicle industries.
My second question is somewhat higher level. When I am cruising on pavement at a consistent speed in classic, the board sometimes feels like it is leaning a bit forward for a few seconds then leaning a bit back for a few. I am trying to stay balanced over the wheel and not rock forward and back, but I am still getting this feeling. Anyone else find this?
Closing note to FM: I strongly encourage you to follow up on this issue. Right now you have a loyal set of users who are willing to take some bruises for the company and the product; we are certainly not looking to damage the company. Help us to help you improve the behavior of the product and make it more attractive to the mass market. If you need some help or ideas, just let me know. Thanks!
ted last edited by
@dj.rknz Ouch....that sucks. Sorry to hear. Hope your arm/shoulder heals soon.
I am on this thread with a couple of similar experiences and a similar request to @onewheel. I hope they do follow up and/or let us begin to collect the data. After each of these "events", it has taken me longer and longer to trust the board and ride again. I love this product and don't want to loose faith in it.
Regarding your questions, the forward/back thing goes away as you ride more. I suggest trying Extreme mode. Seems to be a better (albeit less carvey) ride.
Mynticelx last edited by Mynticelx
I was riding today and rode on and off for about 3 miles. On coming back to my car, I was heading up a slight incline and going what I estimate to be between 6-7mph.. All of a sudden the wheel just locked and I happened to nose dive pretty bad. Thankfully I was wearing my wrist guards, but my arms got shredded pretty bad. I've been trying to analyze what happened. At the end of the day, until OW can give us more recordable data.. We are not going to be able to prove whether it's the rider, board or conditions.
I would happily build an alternative app if FM/OW would open up their data api to bring the sensor data into another app.
Until we get some data collection, we're shooting in the dark as to why these stop the way they do.
OK back to scraping asphalt out of my arms. Cheers
DavidJohn last edited by
I was riding today and rode on and off for about 3 miles. On coming back to my car, I was heading up a slight incline and going what I estimate to be between 6-7mph.. All of a sudden the wheel just locked and I happened to nose dive pretty bad. Thankfully I was wearing my wrist guards,
Same thing here yesterday.. Pretty sure through no fault of my own.. It's a bit disconcerting when these things happen..
badcheese last edited by
The Onewheel is an inherently unstable powered unicycle, not a self-balancing device. All of its sophisticated sensors and controllers merely give it a way to respond to the control input from the rider. It's more stable than a traditional unicycle because the control system can make more precise adjustments to the wheel more rapidly than humans can by turning a crank with their legs, and because the tire is wide. Aside from that, you're still learning to balance a unicycle, but this one goes a lot faster.
New riders have to work hard just to avoid falling, and slow speeds feel fast. Pretty soon they get the hang of balancing. They get the hang of steering. Their confidence improves. Slow speeds feel very slow. They don't feel like beginners anymore.
The stability and power they get from the motor and electronics come with a trade-off: Unlike legs turning a crank, the motor of a Onewheel isn't wired into the rider's brain, so the rider's awareness of the motor's limits is vague. As the rider approaches those limits, the motor loses its ability to improve stability, or even to provide useful feedback to the rider.
At that point, the rider has become the pitiable fool (and let's be honest: We are all this fool) who deliberately rode a powered unicycle in Extreme mode (so that it would be even less able to protect the fool from himself) at a higher speed than he can safely fall...
Anyone with a defective board should certainly have it serviced. The rest of us should understand that we're learning how to ride a powered unicycle at unsafe speeds, and we should adjust our expectations accordingly. It's not safe, nor is it easy, and that's part of the appeal. If you're taking high-speed nosedives, congratulations! You're no longer a beginner! Here's another piece of good news: You're not an expert yet, so you still have lots of cool stuff to learn.
chabis last edited by chabis
@badcheese nailed it. (and this is why you should always wear a HELMET.)
Mynticelx last edited by
@badcheese While I agree with everything you said.. There are some of us that were till in classic mode that this has happened to. I'm not complaining and don't necessarily believe it's the fault of the OW... But just like a unicycler (if that's even a word) would check their mecahnical parts to their bike, I would like the same for the OW. Stream me the data of the ride to understand why these things happen. What as the speed, gyroscope angle, engine rotation, etc. It's not too much to ask to FM to open that up to its riders.
@dj.rknz I would like to know that someone in a senior management or engineering role at FM is reading this thread. Could you please confirm that this is the case? Thanks very much.
@badcheese I'm afraid I must disagree with Mr Cheese on this. A unicycle is an analog device under the direct and absolute control of its rider. Its instability and difficulty are due entirely to the physics of its design. This is the reason it costs $150 instead of $1500.
A OW, on the other hand, is a digitally intermediated, highly-abstracted ride by wire system that depends heavily on behaviorial models embodied in software along with a complex network of sensors and actuators. The fact that it is not self-balancing is immaterial; more material are the facts that the behavior of the system depends almost entirely on the assumptions and judgements of the system's designers, how these are embodied in software and component selection, system testing and revision, heavy end-user input and honest reassessment, and more testing and revision. In this regard a OW is more like a fighter jet than it is like a unicycle. As with the jet, there can be strong perceived pressure on management to "bury the data"; this must be resisted at all costs, particularly with a system as immature as the one we are talking about.
badcheese last edited by
[...] This is the reason [a unicycle] costs $150 instead of $1500.
A cheap unicycle costs $150. What we have is a fancy unicycle. Here's a guy who spent $2800 on his unicycle, and it doesn't even propel itself:
A OW, on the other hand, is a digitally intermediated, highly-abstracted ride by wire system [...] the behavior of the system depends almost entirely on the assumptions and judgements of the system's designers [...] In this regard a OW is more like a fighter jet than it is like a unicycle.
The OW is far from ride by wire. With the right software, we could replace the pilot of a modern fighter jet with a corpse and still fly fancy aerobatic maneuvers. Put a corpse on a OW, and the best you could hope to do with software control is maybe drag it in a straight line for a short distance - assuming the fender kit keeps the corpse from jamming up the wheel, etc. (I'll leave the remaining details as an exercise for the reader...)
Also, keep in mind that pushback just means the wheel speeds up slightly in an effort to get a little bit ahead of the rider's center of balance. The most extreme thing about Extreme Mode is that the pushback is extremely subtle.
bmtka last edited by
@badcheese I think what @dj.rknz meant when he said "ride by wire" is that the rider's physical input does not translate in a direct mechanical fashion to the control of the vehicle, a la a bicycle or unicycle. The input is translated through sensors and processors and then fed as binary information to the motor control unit.
There is a term in motorsport, "brake by wire" which means that a processor controls the brakes based on input from the driver vs. direct mechanical braking where depressing the pedal engages a series of mechanical bits to cause the brakes to be applied. The "feel" and feedback of a "by wire" mechanism isn't going to be as straightforward as one where direct mechanical input is involved. With a vehicle as advanced and new as the OneWheel, there is going to be a lot of room for development to improve all of the things. It would be very nice if FM would release some sort of API to at least monitor the vehicle's data output. Tasker integration would also be incredible, but I don't see that ever happening. I'm afraid that they won't open up even the data output of their proprietary software any because of the looming threat of Chinese knockoffs. You know the Trotter is only the beginning of that shit circus...
But that being said, I agree with you both on your major points.
powder last edited by
Wow: I ate it bigtime in a nosedive scenario on pavement and been on crutches for weeks. I sent the board back to onewheel after they told me it may be defective but now im told its working fine. They are sending it back. This thing did the dive 2 times.
powder last edited by
@smashdunk Expecting the company to respond to any questions that are not real simple and generic is impossible. The have a firewall up to prevent us from talking to anybody except each other re: anything technical, and this toy is technical as hell. All it is is software.