Terrible accident on Ninebot Z10

[Seba]

28 minutes ago, Chriull said:

Actually not.

Ok, "it depends"  But I agree with you, there are actually two situations causing EUC to stop stabilizing itself and consequently the rider. One is a literal cut out caused by protective countermeasures in controller, second is lack of torque reserve caused by mechanisms you've described. Both are resulting in the same and both are abrupt enought to justify use of "cut out" term for both cases.

[Smoother]

10 hours ago, meepmeepmayer said:

I believe EUC BMSes have this disabled specifically so you won't just fall after a millisecond of undervoltage. Early EUCs had hat problem.

 

10 hours ago, svenomous said:

Oh, I thought that EUC BMS still does low-volt cut-off.  So what causes low-battery type failure, then, if not a cut-off?  Is it just that the motor can't get enough amps, but the wheel is actually still powered?

yes that sounds about right.  95 Amps demanded, 90 Amps available, the missing 5 Amps get borrowed from you face as it slides down the road.

Edited May 28, 2019 by Smoother

[Rywokast]

13 minutes ago, Smoother said:

 

yes that sounds about right.  95 Amps demanded, 90 Amps available, the missing 5 Amps get borrowed from you face as it slides across the ground.

my sides 

[Smoother]

7 hours ago, Marty Backe said:

I assure you, it's possible to overlean a KingSong wheel.

The Z10 implements tilt-back just like KingSong. When you reach the max speed it starts to tilt-back to slow you down. You can't turn it off like on Gotway. But if you aggressively lean into the wheel, it only takes a split second to over power the motor.

So just like this Z10 episode, the KingSong wheels will also cutout.

Maybe this is a bit of stereotyping, but I think the typical KingSong owner does not ride their wheel as aggressively as a Gotway or Z10 rider. And this is why cutouts are not reported much for those wheels.

 

7 hours ago, svenomous said:

@Marty Backe, while I do buy the theory that KS riders as a population are statistically less likely to ride aggressively, the problem with forums and YouTube, etc. is that statistics go out the window since it's the outliers who're more likely to be present and reporting their woes.  It would only take a few anomalously aggressive KS riders who suffer a cut-out and post their videos or discussions before the number of reports of such cut-outs would seem just as high as it is for Gotway and others.  So, isn't it also possible that KS is allowing for a larger safety margin that gives more time/torque/power for tilt-back to "succeed" before cutout, thereby contributing to the fact that we see a lot fewer cut-out reports for these wheels?

I'm a King Song rider and I don't think I ride as aggressively as an advanced Gotway rider, but not for the reasons Marty gives.  I ride more conservatively because I am petrified of King Songs agressive tilt back*.  I'm serious, I have been thrown down the road by aggressive tilt back more times than I care to remember.  The thought of hitting it again keeps my riding sane.  Most of these aggressive tilt backs occurred on my KS14C 340WH.  Sometimes it was low battery, and sometimes it was aggressive acceleration to max (low max of 30kph, so easier to hit).  and sometimes a combination of both. Sometimes I managed to stay on, but OMG did I sh..t myself for a second or two!  As a result I have become a master of energy application.  I would never lean into an already fast ride like in the video we are discussing. And as Marty said, if necessary, I reduce my speed to compensate for an impending power spike such as a bump or incline.

Another fear on the small battery wheel was "tilt back to cut out".  The speed sensing routine calls for a tilt back, the tilt back routine calls for more power, the battery says no more power, the wheel say get off....now.  This was the scenario of my worst, and only actual face plant (face? meet road).

 

* aggressive tilt back:when the wheel accelerates fast and suddenly, in order to get in front of a fast moving rider, in order to be able to tilt him backwards to slow down.  Not to be confused with normal tilt back where a gentle acceleration / tilt warns the rider of a situation (speed, power, etc)

 

Edited May 28, 2019 by Smoother

[mrelwood]

 

Quote

It just detects overload condition that could cause catastrophic failure to the controller power stage, so as a countermeasure it turns off all the transistors.

I’m pretty sure that’s not what the wheels do, not since 2016 or even earlier.

Quote

There is no "soft" power limit, it's rather 0-1 situation.

Modern wheels, even the Z10, can’t be this easily be brought to a catastrophic failure mode. (YouTube: Pushing a car with a 100V MSX) The event horizon for both overlean and a cut-out is a sharp line, yes, but what follows after is massively different.

4 hours ago, Seba said:

Both are resulting in the same and both are abrupt enought to justify use of "cut out" term for both cases.

The max tilt angle allowed is 40-45 degrees, I’m pretty sure on all current wheels. Up to that point the wheel will keep delivering all available power. In this case, the same power that was almost sufficient to keep up with the rider.

When the wheel tilts forward 3 degrees at full acceleration, your feet senses that something is off, and by 10 degrees (or after your reaction time) you start to see the film of your life beyond your eyes, lift and woosh your hands all over in hope to have grown wings, and try to hop off in a way that will save your new jeans.

If you just barely pass the overlean limit (battery power or motor back-EMF), the above can last a few seconds (YouTube: ”ACM 45km/h top speed confirmed”).

At an actual cut-out, people constantly mention how they didn’t even have the time to lift their hands in front before hitting the ground. The wheel is instantly dead. Even if your reaction time was zero, you have zero platform to jump off from, and you were already in a forward lean to push thru wind resistance. Perhaps even accelerating on top of that.

In a cut-out you hit the ground in about half a second.

Very different occurences. But the terms are often mixed.

Edited May 28, 2019 by mrelwood

[Nils]

13 hours ago, meepmeepmayer said:

@Igor Rz At what speed was this?

Doesn't look like the Z10's speed limit (45?) to me. Then he should not have been able to overlean the wheel that easily! That looks like a normal sporty acceleration at a still reasonable speed. It should have worked!

Maybe it was some stupid current/whatever limiter throwing the rider off?

The rider commented on the YT video saying "92% of battery and speed about 40-42kmh" FWIW.

[Planemo]

6 hours ago, svenomous said:

@Planemo, I think you're right that in most cases where people overlean, a "cut-out" is not what's happening.

I'm glad I am not the only one.

To be honest I find having discussions like this terribly difficult on the forum. People go off on all tangents, talking about respecting the wheels limits, how there is only so much power, only so much battery, dont lean heavily when near top speed, learn how the wheel works etc etc. and seem to miss/avoid the actual point in question.

I know how to respect a wheel, I havent fallen from an overlean yet and I know not to take the pi$$ when I am already motoring on the thing but I was simply trying to find out why a wheel should shut down just because it cant spin any faster. Point being, a Tesla car doesnt just turn off the entire propulsion system when it reaches top speed.

I totally get that an euc would lose its ability to balance (hence initiate an overlean), but not shut down.

 

[Marty Backe]

46 minutes ago, Nils said:

The rider commented on the YT video saying "92% of battery and speed about 40-42kmh" FWIW.

Good info. Well, that's near the upper limits for the Z10. That speed is the approximate tilt-back speed at 80-percent battery.

[meepmeepmayer]

2 hours ago, Nils said:

The rider commented on the YT video saying "92% of battery and speed about 40-42kmh" FWIW.

Well then, it really was an overlean! Case closed. Thanks!

 I would have guessed he crossed the grass at no more than 30kph from the video.

[ED209]

@meepmeepmayer, it certainly looked like 30kph, this guy should get himself a proper wheel, If he wants to ride like that. 

Edited May 28, 2019 by ED209

[meepmeepmayer]

2 hours ago, Planemo said:

but I was simply trying to find out why a wheel should shut down just because it cant spin any faster.

Because there is no reason not to shut down.

If the wheel is at its limits or already tilted forwards enough very close to its limits, it's already too late. The point is: the rider will fall, no way around it. I see two arguments for switching it off then:

• To protect the electronics from possibly being damaged by some extra spikes.
• To not add more energy to what is already a guaranteed crash.
  Imagine leaning forward heavily at top speed. You know you will fall, the wheel can't catch up. But the wheel just goes on blindly and adds another 5kph to your speed while you're in limbo. Not good! You crash at faster speed. Maybe it even pushes you, grating along the ground in super unlucky circumstances. And right after you have fallen, maybe it runs you over for extra humiliation

So a wheel should relatively quickly switch off at the earliest point it knows it's already too late. That only makes sense. [Whether the specific manufacturer's implementation of this is good or too conservative or ... is another question entirely, but the principle stands.]

No point in investing in a doomed endeavour. My interpretation. Maybe there are other technological reasons.

[seage]

7 hours ago, Smoother said:

I'm a King Song rider and I don't think I ride as aggressively as an advanced Gotway rider, but not for the reasons Marty gives.  I ride more conservatively because I am petrified of King Songs agressive tilt back*

Holy shit, yes, thisssss! Back when I was on my 16s, i went on a group ride. 50km ride for the first time. We had a charge break and all, but coming back onto that final stretch, my battery was at 23%. Lowest it had ever been. I had already slowed down a lot in anticipation of it starting to tilt, just no one told me it was going to try to kill me itself. I was riding around 20kmph and accelerated up to i think 25kmph and it just shot into a tiltback EXTREMELY hard. I wasn't pushing hard, i didnt lean into it, it didnt gradually tilt, matching my acceleration, nope, just ripped me backwards. It scared the hell out of me and almost threw me backwards off the wheel. My heart damn near exploded. I never want to feel tiltback again, haha. 

[kasenutty]

I've said it before and I'll say it again, I much prefer tiltback to tiltforth. 

[svenomous]

Well, wow.  I left the thread yesterday and have come back to a lot of new posts that have mostly decreased my level of understanding of the technology factors, due to contradictions.  I think I'll just stop trying to say anything about BMS, mainboard, torque, etc., and just fall back on what is (I believe) not debatable: a self-balancing wheel must stay under the rider; a rider whose CG is in front of or behind the wheel's CG forces the wheel to move to maintain equilibrium, which is how we get the graceful glide of EUC riding; a wheel cannot decide what speed it goes at, as that speed is determined by the force vectors involved, and is always equal to exactly what will keep the wheel in equilibrium; a wheel's ability to "stay under" a rider is limited by several technical factors that have been discussed on this thread; when a wheel's ability to stay under a rider is exceeded, the immediate result is loss of equilibrium and a very graceless fall.  So: whether it's due to torque limits, battery percentages, power spikes like bumps/holes/ruts, or limitations of mainboard power components, one should appreciate that fundamentally an EUC in failure mode will dump the rider, unlike an inherently stable vehicle with more wheels, which can continue right at the edge of failure or coast to a gentle stop from a failure.  So: be respectful and cautious, have some sense of where a wheel's "edge of envelope" lies, and don't assume the wheel safety features will save you if you're being aggressive enough in your riding.

[Michael Tucker]

I agree with @Planemo, I wonder if these wheels are designed properly too. In my mind balance, acceleration, and speed are three separate performance attributes that should be independent, with the priority going to balance (level pedals). If a rider leaning forward is nothing more than an input telling the wheel to accelerate, shouldn't there be a safe predetermined limit of speed(rotation of the wheel) where any further input telling speed to increase would simply be ignored, with a greater focus on balance rather than matching faster acceleration. If my wheel was controlled with a remote like an eBoard instead of leaning, where I simply pressed a button for acceleration, would the wheel accelerate too fast or reach too high a speed simply by pressing the button? If so, would that be poor engineering or me pressing the button incorrectly?! 

I know the wheel can adjust its pedal angles independently from the riders input (KS sharp tiltback), so spinning the wheel slower while maintaining level pedals should be possible. There does not seem to be any need for fast dips, cut-offs, over-leans or even sharp tiltbacks. This problem will definitely be solved in future wheels, as there is simply too much guesswork with safe operation of modern wheels. Riding one of these things at speed is more akin to piloting an aircraft than any other device with a wheel. Whats my speed? How fast am I accelerating? What battery level am I at? What's the slope of this hill? How hot is the weather? How old is my wheel/battery pack? Whats my PSI? What temperature is my mainboard and motor wires? What firmware am I running? How windy is it? Is it raining? Am I restricted from riding here...

[svenomous]

@Michael Tucker, no.  Just about everything you stated as "should be possible" is not.  If you had a stable platform (sitting on multiple wheels) and a separate "tilting motor" you could do the things you propose as possible, but you don't.  Instead you have a platform suspended on an axle, and a single motor that rotates the wheel at varying accelerations/speeds to achieve all the performance attributes you called out as needing to be separated: balance, acceleration, and speed.  Therefore, these attributes are intimately connected.  A particular pedal angle being maintained (and the rider being kept from falling) is due to accelerometers/gyros in the wheel's case constantly monitoring its longitudinal angle, and as it moves by micro-degrees motor power is varied to correct.  It's like a circus balancing act where the juggler puts a plate on a stick and the stick on his forehead, then moves continuously to keep that plate from falling off the stick.  The way pedal angle changes is by accelerating or decelerating a tiny bit, and then using the new resulting angle as a new reference for "equilibrium."

If the rider puts CG forward and the wheel does not immediately accelerate to maintain the angle/tilt, the system immediately loses equilibrium and the rider falls.  No technological advancement will change this fact.  It's physics, not technology.  This is the same as the juggler example: the juggler can't decide to just not move past a certain point while balancing that plate...if he slows or stops the plate immediately falls.  The juggler can move rapidly ahead of the plate to stop its movement, and even encourage it to move back the other way, but unfortunately the EUC rider (unlike the plate) has a will and a way to change her CG.  As with the juggler and the plate, tiltback is the wheel accelerating slightly to get its own CG ahead of the rider's CG, which causes the case/pedals to tilt back, in an attempt to force the rider's CG to get behind the wheel so the wheel can stop accelerating or slow.  Physics can't be cheated, but a rider can be strongly encouraged to move her CG backward in this manner.

[meepmeepmayer]

51 minutes ago, Michael Tucker said:

In my mind balance, acceleration, and speed are three separate performance attributes that should be independent

In reality, these are the same. That's just what self-balancing entails.

If a wheel tilts, it accelerates into that direction until the tilt is gone again (or won't go away and becomes too big, which is interpreted as a crash). That's literally all a EUC does. You tilt it, it tries to stop you.

51 minutes ago, Michael Tucker said:

shouldn't there be a safe predetermined limit of speed(rotation of the wheel) where any further input telling speed to increase would simply be ignored

Then the rider would lean forward but the wheel wouldn't follow, so the rider falls off. A EUC can either keep balancing (following the rider's center of gravity) or not (then the rider falls off). There's no other option.

51 minutes ago, Michael Tucker said:

I know the wheel can adjust its pedal angles independently from the riders input (KS sharp tiltback), so spinning the wheel slower while maintaining level pedals should be possible.

No. First of all it would have to speed up to overtake the rider and bring his center of gravity behind the wheel. Only then it can apply tiltback so the rider can't lean forward again and must slow down (center of gravity stays behind the wheel for a while).

One can imagine some advanced pedal tilt trickery to do stuff (like some emergency behavior), but it all has to start with the wheel overtaking the rider. That can't be cheated. So you can not do this at the power limit, only earlier. And since there seem to be huge variations in power demand, presumably you can either be crippingly conservative (like forced tiltback at half the technological top speed) or will always be able to overlean a wheel in some way if you really try.

Self-balancing works just like you: if you stand on the ground and lean forward, you either have to catch yourself by stepping forward with a foot, or you'll fall over. No way around it.

Edited May 28, 2019 by meepmeepmayer

[seage]

1 hour ago, kasenutty said:

I've said it before and I'll say it again, I much prefer tiltback to tiltforth. 

I don't want tilt back or forth. I just ride within the safety margins and monitor my battery level, lol. No tilts for me, this aint a seesaw. 

[Chriull]

1 hour ago, Michael Tucker said:

In my mind balance, acceleration, and speed are three separate performance attributes that should be independent

As both posters before wrote, they are and cannot be "different" and independent. Here another explanation.

Acceleration is by definition the change of speed over time. That's a physical definition. So these two values are strongly connected and can (easily) be computed one from another.

And balancing is just the speed change (=acceleration) in a way that the rider does not fall of the wheel.

So the question arises - what causes the acceleration of the wheel? That's the force (torque) - its even directly proportional, as the motor current.

And here follows another connection: the absolute maximum torque is availble at standstill, 0 at a bit above lift cut off speed and inbetween linearly decreasing.

The last point is the direct proportional connection between lift cut of speed and batter voltage.

... That's all one needs to know to understand the overlean 

PS.: Another limit is "artificially" introduced vy the firmware - the current is limited to a max value, so that the wires, motor coils and mosfets don't fry. As current is proportional to torque/acceleration this imposes another torque/acceleration limit. Which is in effect only for slower speeds, as the torque available from the motor decreases with speed and hence this motor limitation once becomes the limiting factor.

Edited May 28, 2019 by Chriull

[erk1024]

Watching the clip, there were other risky moves: going from dirt across thick grass and back onto sidewalk at a shallow angle. Grass can hide sand, holes, sprinklers, bumps, and it's uneven in thickness. The level of the dirt can be lower than the cement and it's hard to tell, so there could be a lip getting back onto the cement. Hitting the lip at a shallow angle could cause the wheel to skid sideways. Grass can be wet, especially at night. I wouldn't be too cavalier about crossing a patch of grass at speed, at night.

Edited May 28, 2019 by erk1024

[Damarafaka]

On 5/27/2019 at 3:52 PM, Marty Backe said:

It's amazing the lack of knowledge that I continue to see out there on how our wheels work and their limitations. 

I have always had a deep respect for Marty's opinions and precise technical knowledge of eucs as well as his focus on security and even his conservative riding style even though I prefer, as he usually calls it, a more aggressive way of riding. Once again I agree with his assessment of over-lean as the main cause of the cut-off, not so much the idea that the lack of knowledge or recklessness not respecting euc limitations as the reason for suffering cut-offs. What I am trying to say is that no matter how much theoretical information we have, sometimes it is necessary to experience this issue to truly understand it, or at least get a better understanding. In this case it seems the rider isn't going fast enough to cause such a cut-off but one has to take into consideration the amp peaks and as Marty explained very well, he was accelerating hard already moving fast with the wheel probably still bouncing around some after landing on the concrete again, as well as the riders weight!

I'm no expert in voltages, currents, battery power, etc. I just ride fast, too fast, that's my problem. For many we are a bad example for the community, even a danger for fellow riders. I believe that it is also the raw experience of the falls and their riders and videos what allows one to go forward, even higher. I ride a hacked Z10 (limited to 61kmh) in traffic every day and whenever I can I run speed laps in a controlled environment. Maybe I've been lucky with my falls but I'm not afraid of falling. Of course a high speed crash can be fatal but as most of us have found out, trying to not hit a small kid or a stupid fall against a curb at 10kmh cause the worst injuries. The Z10 like most eucs works with something similar to the 80% gotway power limit: near the speed/torque limit, amp peaks (either by bouncing off of bump or/and overleaning will cause TEMPORARY cut-offs! Only crossing these limits can one understand what a cutoff really is and the performance of our wheel, always different. Not only the "usual" cutoffs, far away from the absolute high speed when re-accelerating or after obstacles/amp peaks - the wheel remains in a "safe" speed zone - sometimes even unnoticeable (we hold the pothole accountable for what was really a cutoff), but even the major ones caused by still being in a "slight" overlean when crossing the max speed. After the cut-off the Z10 comes back to life as soon as it feels all parameters are under certain values again. That means that if we can anticipate with our riding posture (more kidney leaning), remain calm taking weight off the pedals but keeping tension in our legs and "riding" the moment, a fall CAN be avoided. I have also the logs of a cut-off after re-leaning at 60kmh, dropping instantly down to around 46kmh before I felt the pedals come back to life and miraculously staying on. 

My only intention is protest the sometimes condescending attitude towards the lets call them more adventurous risk takers and acknowledge the practical information they can bring to the table with what seems at first only a bad or unruly example. Maybe risking my ass I can also help someone come out smiling, that I can bet you, after knowing that cutoffs don't always have to end with a faceplant and actually recover from one.

2019_05_19_18_09_01 cutoff.csv

[Marty Backe]

37 minutes ago, Damarafaka said:

I have always had a deep respect for Marty's opinions and precise technical knowledge of eucs as well as his focus on security and even his conservative riding style even though I prefer, as he usually calls it, a more aggressive way of riding. Once again I agree with his assessment of over-lean as the main cause of the cut-off, not so much the idea that the lack of knowledge or recklessness not respecting euc limitations as the reason for suffering cut-offs. What I am trying to say is that no matter how much theoretical information we have, sometimes it is necessary to experience this issue to truly understand it, or at least get a better understanding. In this case it seems the rider isn't going fast enough to cause such a cut-off but one has to take into consideration the amp peaks and as Marty explained very well, he was accelerating hard already moving fast with the wheel probably still bouncing around some after landing on the concrete again, as well as the riders weight!

I'm no expert in voltages, currents, battery power, etc. I just ride fast, too fast, that's my problem. For many we are a bad example for the community, even a danger for fellow riders. I believe that it is also the raw experience of the falls and their riders and videos what allows one to go forward, even higher. I ride a hacked Z10 (limited to 61kmh) in traffic every day and whenever I can I run speed laps in a controlled environment. Maybe I've been lucky with my falls but I'm not afraid of falling. Of course a high speed crash can be fatal but as most of us have found out, trying to not hit a small kid or a stupid fall against a curb at 10kmh cause the worst injuries. The Z10 like most eucs works with something similar to the 80% gotway power limit: near the speed/torque limit, amp peaks (either by bouncing off of bump or/and overleaning will cause TEMPORARY cut-offs! Only crossing these limits can one understand what a cutoff really is and the performance of our wheel, always different. Not only the "usual" cutoffs, far away from the absolute high speed when re-accelerating or after obstacles/amp peaks - the wheel remains in a "safe" speed zone - sometimes even unnoticeable (we hold the pothole accountable for what was really a cutoff), but even the major ones caused by still being in a "slight" overlean when crossing the max speed. After the cut-off the Z10 comes back to life as soon as it feels all parameters are under certain values again. That means that if we can anticipate with our riding posture (more kidney leaning), remain calm taking weight off the pedals but keeping tension in our legs and "riding" the moment, a fall CAN be avoided. I have also the logs of a cut-off after re-leaning at 60kmh, dropping instantly down to around 46kmh before I felt the pedals come back to life and miraculously staying on. 

My only intention is protest the sometimes condescending attitude towards the lets call them more adventurous risk takers and acknowledge the practical information they can bring to the table with what seems at first only a bad or unruly example. Maybe risking my ass I can also help someone come out smiling, that I can bet you, after knowing that cutoffs don't always have to end with a faceplant and actually recover from one.

2019_05_19_18_09_01 cutoff.csv 39.12 kB · 0 downloads

Thanks for the nice words 

Personally, I'm never bothered by the speed freaks (you ) or people who use our wheels on the edge. I love it all. I know there are people who enjoy seeing me ride wheels up extreme hills yet have no intention of doing so themselves. Likewise, I like to see the wheels pushed in ways that I'll never try. As a community we all grow by seeing and understanding what they can do.

So please, continue risking your ass and reporting back periodically that you survived 

[mrelwood]

11 hours ago, Michael Tucker said:

spinning the wheel slower while maintaining level pedals should be possible.

You already got the answers why this can’t happen. I think that the amazing balancing capabilities of our wheels disturb the thought process on how stuff works.

An euc is an axle (shell and pedals attached) and a spinning wheel. Nothing more. The only thing an EUC can do at any time is spin the wheel. Nothing else.

What is amazing is that it monitors the tilt position and adjusts the speed of the wheel several hundred times every second. That makes the pedals feel like a stable platform. But they are not any more solid than when the wheel is powered off. All the stability comes only from accelerating the wheel with incredible precision. Nothing more.

These are incredible devices. But laws of physics still have the priority.

[Chriull]

4 hours ago, Damarafaka said:

 Once again I agree with his assessment of over-lean as the main cause of the cut-off, not so much the idea that the lack of knowledge or recklessness not respecting euc limitations as the reason for suffering cut-offs. What I am trying to say is that no matter how much theoretical information we have, sometimes it is necessary to experience this issue to truly understand it, or at least get a better understanding

I don't like the overlean is mostly called cut-out, although it feels like one it is not. There is nothing cutting off, just the wheel delivering less torque.

Quote

...

The Z10 like most eucs works with something similar to the 80% gotway power limit: near the speed/torque limit, amp peaks (either by bouncing off of bump or/and overleaning will cause TEMPORARY cut-offs! Only crossing these limits can one understand what a cutoff really is and the performance of our wheel, always different. Not only the "usual" cutoffs, far away from the absolute high speed when re-accelerating or after obstacles/amp peaks - the wheel remains in a "safe" speed zone - sometimes even unnoticeable (we hold the pothole accountable for what was really a cutoff), but even the major ones caused by still being in a "slight" overlean when crossing the max speed. After the cut-off the Z10 comes back to life as soon as it feels all parameters are under certain values again.

The "limit" is nothing like 0/1 - that's why i wrote above i do not like the term "cut-off". Reaching the torque limit does not turn the motor/EUC/or whatever off and once one is inside the limit turn it on again. (Although it can be seen/described this way as a first very rough approximation)

In reality one "never" drives with an EUC with a fixed torque at a constant speed. If one would have the exact data with a good timely resolution the points in a diagram would look like a "cloud" - by the self balancing the wheel is always in some "torque range" and "speed range" changing a couple of 100 times per second. 

So if one nears the limit, this "cloud" starts to touch the limit with some of its periphery. At this "outliers" the motor cannot deliver the full torque, but as these are just some "seldom" points and the majority of points in the cloud are well within the limit where enough torque is available, everything is still "ok" but not as perfect selfbalancing as if the whole cloud would be within the limit. This is imo felt as the "pedals getting weaker" - or one gets a bit a bad gut feeling of "missing support".

The nearer one comes with the "center of the cloud" to the limit, the worse the self balancing of the wheel gets.

Quote

That means that if we can anticipate with our riding posture (more kidney leaning), remain calm taking weight off the pedals but keeping tension in our legs and "riding" the moment, a fall CAN be avoided.

Yes. The slower one nears the limit, the more one "anticipates" it and the more experience one has the more probable/easier it gets to prevent a fall. As this is just an overlean and no cut-off that's the way it goes

Human beeings have an amazing balancing capability - it has to be trained but then one can do incredible stuff!

But if one just wildly accelerates into the limit while leaning forward as much as one dares there will be no way out but flying...

Or if one hits a big enough pothole driving with not enough reserves - then nothing can help one to stay with the wheel...

Quote

I have also the logs of a cut-off after re-leaning at 60kmh, dropping instantly down to around 46kmh before I felt the pedals come back to life and miraculously staying on. 

Here your logs shown as Current (== torque) over the speed you've driven with the limit line in the graph. The limit line is a bit unprecise (i don't have enough data and every wheel is a bit different) - but it shows about where it could be

 

And a zoom in in the upper interesting part:

Here one can imagine how this "cloud" could be - especially in the area before coming near the limit, the points are zig-zagging around as the wheel performs the stabilization. This are just the points reported from the wheel and logged some ~5 times a second. And the Z10 is very "lazy" reporting and pauses inbetween. And it does some strange "rastering" as seen on the artificially looking straight lines... 

But internally this points are calculated this some 100 times a second and the motor drivers adjusted accordingly.

Quote

My only intention is protest the sometimes condescending attitude towards the lets call them more adventurous risk takers and acknowledge the practical information they can bring to the table with what seems at first only a bad or unruly example. Maybe risking my ass I can also help someone come out smiling, that I can bet you, after knowing that cutoffs don't always have to end with a faceplant and actually recover from one.

As you wrote "running speed laps in a controlled environment" is for sure a great way of experienceing ones wheel and learning about its limits, dealing with it and training the own body/brain to help balancing!

But one will have to take some crossing the limits and flying off the wheel while doing so.... which trains the ability to fall more safely....

Edit: PS.: The whole thing above is just easily experiancable with a wheel with disabled tiltback, like your modified Z10 or some GW. With a Kingsong (or Inmotion) one has no chance to come slowly and controlled to the limit

As you see in the graph above - with a normal Z10 one is just able to drive in about the "green" area, then tilt back kicks in...

Edited May 29, 2019 by Chriull

[Seba]

All this interesting discussion has led me to consider adding the possibility of adding data from phone accelerometer and gyroscope to WheelLog log files. In this way the wheel parameters would be complemented with the information about forces acting on the rider. Of course, a phone is not usually rigidly attached to the rider, but it could still be very valuable information in the analysis of such events.