Why do wheels "cut out?"

[Yellowjacket]

It seems "cut out" is a much-feared and not uncommon danger on some EUCs. Why?

Is it not possible to design software / firmware / board hardware such that a wheel would remain stable while ramping speed down to a few kilometers per hour? (Assuming no total board failure).

Is our sport so small that this kind of safety feature is not worth the effort by manufacturers? Would the compromises in performance be so great as to make a "ramp down" unacceptable? Is the need to have the wheel "turn off" in some circumstances logically incoherent with a "ramp down" feature?

Will the "engineers" help my understanding?

[DjPanJan]

You mean CUT-OUT = sudently lost power = Hardware fail or bad mainterrance. Some type ultrasafe EUC what is double bufered mean price over 10 000 eur/dollar if ouy want keep range and weight curent EUCS in early 2021.

More often is just Overlean = rider simply push EUC over maximal possible "power" and is no more energy/power to keep ballance. 

[.....]

electronics can fail, its inevitable. Batteries can fail, its inevitable. Quality is very dependant on choice of parts. Companies make compromises to reach certain price points. Until the market is willing to pay out the nose, we will continue to see failures more often than if we were to be riding on more expensive vehicles with less compromises in quality. A BIG step in that direction, would be to patron a company with a responsibility to the customer and being held liable for failures. The EUC chinese market doesnt have those worries.

Edited February 8, 2021 by ShanesPlanet

[Chriull]

18 minutes ago, Yellowjacket said:

It seems "cut out" is a much-feared and not uncommon danger on some EUCs. Why?

As @DjPanJan wrote you mean an overlean - does normally not feel too different but there is no hardware or software failure involved.

18 minutes ago, Yellowjacket said:

Is it not possible to design software / firmware / board hardware such that a wheel would remain stable while ramping speed down to a few kilometers per hour? (Assuming no total board failure).

No.

If one leans foward the wheel has to accelerate to keep one balanced.

If the wheel ramps the speed down (==deceleration) one falls on ones face - that's exaxtly the faceplant/overlean one does not want.

18 minutes ago, Yellowjacket said:

Is our sport so small that this kind of safety feature is not worth the effort by manufacturers? Would the compromises in performance be so great as to make a "ramp down" unacceptable?

As written above it's not unacceptable performancewise but make one overlean and faceplant earlier...

18 minutes ago, Yellowjacket said:

Is the need to have the wheel "turn off" in some circumstances logically incoherent with a "ramp down" feature?

The wheel does not turn off in the case of an overlean - it just does not accelerate anymore enough to keep the forward leaning rider balanced. In "high acceleration" scenarious this just feels like a cut out.

18 minutes ago, Yellowjacket said:

Will the "engineers" help my understanding?

It's all about a "torque equality". A rider forward inflicts the torque in one rotational direction. Balancing of an EUC works by the motor generating a torque in the counterdirection by acceleration. So the pedals stay straight.

If the EUC "overaccelerates" the pedal tilt up, if the EUC can't accelerate (provide enough torque) enough one tilts/falls forward...

Easy, isn't it?

 

[Yellowjacket]

40 minutes ago, Chriull said:

The wheel does not turn off in the case of an overlean - it just does not accelerate anymore enough to keep the forward leaning rider balanced. In "high acceleration" scenarious this just feels like a cut out.

It's all about a "torque equality". A rider forward inflicts the torque in one rotational direction. Balancing of an EUC works by the motor generating a torque in the counterdirection by acceleration. So the pedals stay straight.

If the EUC "overaccelerates" the pedal tilt up, if the EUC can't accelerate (provide enough torque) enough one tilts/falls forward...

Easy, isn't it?

 

So a wheel "cut out" is just a matter of rider error? Asking for more than the wheel has available? And there is always a warning when this condition is immanent? Thank you for your explanation.

[Chriull]

17 minutes ago, Yellowjacket said:

So a wheel "cut out" is just a matter of rider error?

Normally yes. Then it's an overlean.

Of course "real cut outs" with hardware can and do happen.

17 minutes ago, Yellowjacket said:

And there is always a warning when this condition is immanent?

Unfortionately no.

The GW 80% alarm is often/mostly a great warning.

The fixed speed tiltback of the other wheels does not warn/help from high accelerations at higher speeds.

KS has a a "newer" value - inverter load, whicg generates an 88% warning. That's a much more profound value used for warning but unfortionately much too late for higher accelerations.

Some dynamic use, sensefull and reliable warning is still missing fir EUCs.

Don't know why, but it seems 

1 hour ago, Yellowjacket said:

not worth the effort by manufacturers?

...

[Yellowjacket]

Then when I buy my next wheel (suspension, less than 20 kg, built with with quality control and customer in mind) I'd like the wheel to give awarning at 70 percent, another unambiguously at 80 percent, both values alterable by the rider.

[.....]

31 minutes ago, Yellowjacket said:

Then when I buy my next wheel (suspension, less than 20 kg, built with with quality control and customer in mind) I'd like the wheel to give awarning at 70 percent, another unambiguously at 80 percent, both values alterable by the rider.

If you EVER find such a machine, let us know... Most current wheels on the market work well with eucw so you can make your own user alarm settings to cater to your riding style. I view 'cutouts' as a hardwre failure. Either rider induced by riding beyond the limits of the circuit and causing fail, or by merely bad parts. "overlean" is usually NOT an equipement failure, it is simply when a rider is asking more power than the batteries can provide. "overlean" is rarely a fault of the hardware, as EVERY piece of equipment has a design limitation, and overleans are simply exceeding them.

Edited February 8, 2021 by ShanesPlanet

[mike_bike_kite]

23 minutes ago, Yellowjacket said:

Then when I buy my next wheel (suspension, less than 20 kg, built with with quality control and customer in mind) I'd like the wheel to give awarning at 70 percent, another unambiguously at 80 percent, both values alterable by the rider.

Suspension & less than 20kg means you might be waiting a while. In the meantime you should try the alarms on the phone app EUC World. It has all sorts of alarms you can set. 

[Ádám Szitás]

I think cutouts are overrated. Wheels these days very very rarely just shut off and overleans are hard to achieve, if you are even a little self aware.

[meepmeepmayer]

Warnings are complicated. On the one hand, manufacturers clearly could implement much better warnings. On the other hand, reliable warnings are really hard to get right.

-

As for "why"...

Our electric motors work like this:

• A given motor has a fixed maximum speed.
• That speed depends on the voltage (which you see as battery percentage) applied to the motor. Higher voltage, higher max speed. 100V wheels are faster than 84V wheels using the same motor. A full battery allows for a higher max speed than an empty battery.
• The motor torque falls towards zero at the max speed.

Torque is what keeps you upright and balanced. So you need to stay below the max speed, with a decent safety margin. Running out of torque means an overlean crash.

Two things reduce torque:

• Going fast. As said above, the closer you are to the max speed of your motor, the less spare torque you have as a safety margin.
• Accelerating (drawing power from the battery). That drops the battery voltage, which means your motor's max speed suddenly falls. The harder you accelerate, the worse the voltage drop, the more your max speed is temporarily lower.
  (Also you could simply temporarily draw more power from the battery than it can provide at all. That is technically also a voltage drop - to nothing at all, so at any speed you are faster than you can be.)

So you have a torque budget. The faster you ride, the smaller it is. And accelerating eats into the budget temporarily, while you accelerate. If at any point you get over the limit, that's an overlean.

As you might guess, the easiest way to get an overlean is to ride very fast and then accelerate hard. That's how all the "famous" overleans happen, where the crazy riders go 50mph on NYC streets and then accelerate a bit more and... well, boom. To be fair, this is how they find out what they can and can't do on a given wheel

You could have an overlean by accelerating crazy hard at a more normal speed, but in practice that rarely happens, and only on wheels with small batteries (think V8). Small batteries distribute the power draw among less battery cells, so the voltage drop is higher, and also the battery might simply be overwhelmed more easily. A modern big battery wheel (think V11) is virtually impossible to overlean from anything but too high speed. A V8, you might be able to overpower it at 20kph if you really try and weigh 120kg.

To summarize:

• If you just go fast enough, you can overlean any wheel. In theory and practice, no way around it. The motor has a max speed, and self-balancing entails that the wheel can't outvote you if you lean it. So if you keep leaning, it must go faster and faster and faster... till it can't.
  In reality, tiltback is a great way to stop this. The wheel will tilt back, it is uncomfortable and unergonomic to stand on the tilted pedals (but you could!), so you automatically slow down to make the pedals flat and comfortable again. In practice a wheel can quite effectively slow you down if it wants to.
• If you just accelerate hard enough, you can (mostly theoretically) overlean any wheel.

So: high speed, hard acceleration, low battery charge, and things that weaken the battery (= cold battery) are what kills you. Especially in combination.

Now to the warnings:

• All the various "80%" and "high speed" alarms just mean... you're at 80% (or whatever) of the max motor speed for the given voltage. Voltage is battery charge, so that's why these alarms appear at lower speeds the less battery charge you have.
  Such alarms are easy in theory and easy to implement. You know the max speed for a given motor and battery voltage, no uncertainty here. Choose your 80% or 75% or 70% of that speed, make the wheel beep/tiltback then, done! If you accelerate, the voltage drops, and the alarm automatically kicks in at the now lower max speed. Low battery? Means lower voltage, means the alarm kicks in at a lower speed. And so on. It automatically adapts.
  In theory, that's the only alarm you will ever need. It will always sound if you get too close to low torque. Problem solved! (No.)
• Now the problem is: accelerating.
  You can accelerate hard, and you might be on the ground before the alarm had a chance to warn you. That could always happen if you already in the low margin regime (= you're riding "too" fast).
  How big do you want your safety margin for accelerations to be? The harder the accelerations you want to cover, the more you need to reduce the alarm speed. So your 80% alarm suddenly becomes a 60% alarm or 40% alarm?
  Do you want to add any other kinds of alarms, for example one that reacts to a voltage drop, not the voltage itself?
  It's a delicate question when the alarms should sound, how they should work, etc. There's no easy formula here. And worst of all, it's a trade off between safety and your wheel's capability (the all-important top speed!).
  So this requires effort, experimenting, real world experience to get it right for various situations and wheels. The manufacturers are getting better (e.g. pedals getting soft when you draw a lot of power to discourage further hard acceleration) but I'm sure they could do a much, much better job here.

In reality, all this doesn't matter too much. An overlean from anything but going too fast is extremely rare. If you always stay 10kph below the beep speed of your wheel, have a big battery, ride with a charged battery, make sure the battery isn't really cold, and are mindful that the faster you are the softer your accelerations should be... you will not get an overlean unless you intend to

Side note: a 20kg super powerful wheel won't happen (soon) precisely because you need a big battery for that (safeguard against voltage drop from hard accelerations), which increases the weight.

Edited February 8, 2021 by meepmeepmayer

[AtlasP]

@meepmeepmayerThis is quite possibly the single-best post I've ever seen on this forum. Certainly within the top-five or top-ten. (Side note: we need a pinned reddit-style, best-post-of-the-year nominations thread with voting at the end of the year.) It should (also) be re-posted/pinned somewhere more prominently where more new riders can find it and people can easily direct others to it when the subject comes up.

Edited February 9, 2021 by AtlasP

[Rich Sam]

Take with a grain of salt It's been like a million years since my electronics classes, but it was my understanding that Power = Voltage*Current

84 volt battery system would have to draw on more current to equal the power of a 100 volt system.  Theoretically they could be a very similar experience if the 84 volt system was rated for high current rates.

I think the big advantage of 100 volt system is it requires less current for equal power in comparison to an 84 volt system making it more efficient, but the downsides are it requires more cells in parallel to up the voltage, would require more cells for same capacity if I'm not mistaken.

[meepmeepmayer]

@AtlasP Thank you Half of writing it down is to understand it better myself. I don't envy the manufacturers having to deal with this as well as the real world behavior of all the (cheap) electronics.

24 minutes ago, Rich Sam said:

Take with a grain of salt It's been like a million years since my electronics classes, but it was my understanding that Power = Voltage*Current

84 volt battery system would have to draw on more current to equal the power of a 100 volt system.  Theoretically they could be a very similar experience if the 84 volt system was rated for high current rates.

I think the big advantage of 100 volt system is it requires less current for equal power in comparison to an 84 volt system making it more efficient, but the downsides are it requires more cells in parallel to up the voltage, would require more cells for same capacity if I'm not mistaken.

The motor's max speed is directly proportional to the voltage. A 100V wheel can be literally 20% faster than a 84V wheel with the same motor (24s to 20s is 20% more). Your 80% alarm speed would always be 20% higher, so you do get the speedup in reality.

It certainly helps that, for a given power draw, the current is lower on a high voltage wheel. That can also help if one of the warnings or firmware limitations is at a fixed maximum current.

But you can be literally faster on a higher voltage.

Not sure what you mean by capacity, but a cell has a certain capacity (in Wh), no matter how the battery is configured. Same number of cells = same battery size. A higher voltage means you have less flexibility in making different battery sizes, because for example you have to go in multiples of 24 cells (100V) instead of 20 cells (84V).

And you could argue, for the same battery size, more serial cells (higher voltage) could mean less parallel cells, so less current possible and a higher voltage drop. But in terms of power draw, the higher voltage balances that out. And in reality, they usually just make the batteries 20% bigger in 100V wheels compared to 84V wheels, for example the V11 (84V, 1500Wh) and Begode 100V 1800Wh wheels are both 4p, so same current and voltage drop.

Edited February 9, 2021 by meepmeepmayer

[Rich Sam]

2 hours ago, meepmeepmayer said:

@AtlasP Thank you Half of writing it down is to understand it better myself. I don't envy the manufacturers having to deal with this as well as the real world behavior of all the (cheap) electronics.

The motor's max speed is directly proportional to the voltage. A 100V wheel can be literally 20% faster than a 84V wheel with the same motor (24s to 20s is 20% more). Your 80% alarm speed would always be 20% higher, so you do get the speedup in reality.

It certainly helps that, for a given power draw, the current is lower on a high voltage wheel. That can also help if one of the warnings or firmware limitations is at a fixed maximum current.

But you can be literally faster on a higher voltage.

Not sure what you mean by capacity, but a cell has a certain capacity (in Wh), no matter how the battery is configured. Same number of cells = same battery size. A higher voltage means you have less flexibility in making different battery sizes, because for example you have to go in multiples of 24 cells (100V) instead of 20 cells (84V).

And you could argue, for the same battery size, more serial cells (higher voltage) could mean less parallel cells, so less current possible and a higher voltage drop. But in terms of power draw, the higher voltage balances that out. And in reality, they usually just make the batteries 20% bigger in 100V wheels compared to 84V wheels, for example the V11 (84V, 1500Wh) and Begode 100V 1800Wh wheels are both 4p, so same current and voltage drop.

I thought power is power and did not matter the voltage.  P=V*i

so if your motor needs 2000 watts with 84 volts that would be almost 34 amps

on 100 volt system it only requires 20 amps to feed 2000 watts

Now there may be a max amperage output in cells used for euc to make it 20% faster, I'm not sure.

 

[Yellowjacket]

4 hours ago, mike_bike_kite said:

Suspension & less than 20kg means you might be waiting a while. In the meantime you should try the alarms on the phone app EUC World. It has all sorts of alarms you can set. 

I believe you are right, given the push to ever more powerful / faster wheels. But I think manufacturers are missing a market.

I live part time on a boat which involves lifting my wheel with one hand up and over a saltwater gap between boat and dock. 14 kilos is easy, 20 kilos is doable. Even when not on the boat, lifting a 20 kilo wheel out from behind the driver's seat of my car is doable, but much more than that with one hand and I'm risking fit and finish on wheel and car.

My two year-old InMotion V8 is capable of about 30 kph, and weighs about 14 kilos. Here's the thing: for the riding I do, 40 kph would be plenty. I don't need nor want to do 50 kph — 60 kph on bike paths, city streets, and certainly not off-road. I also don't need to do 50 kilometers between charges. So, it does not seem much of a stretch that my V8 could be improved to do 40 kilometers per hour if it weighed 6 kilo more, spread between motor, batteries and suspension.

Suspension? Trail riding, curb jumping, pothole absorbing — it's good for safety, comfort and wheel longevity.

So, I agree, I'll be waiting a while, but not because of any technical issues. I think a light, well-made suspension wheel of 20 kilos capable of 40 kph and 40 kilometers of range would find a great market. Of race cars it's said you can build a car that has speed, a low price and longevity — pick two out of three. By limiting speed and range, I think such a wheel would be great fun and affordable.

[Tawpie]

The reason an electric motor can make more RPM at higher input voltage is because as a motor spins it also acts as a generator and creates what's called "back EMF (back-electromotive force)" that in essence counters (cancels out) the voltage being input to the motor. The faster the motor turns the higher the value of the back EMF (the back EMF is proportional to motor RPM). As the motor spins faster and faster, the input-voltage-cancelling back EMF also increases and eventually reaches a point where the motor effectively has very little remaining voltage to work with and it can't go faster.

So if you are feeding the motor 84V and it reaches the RPM that is limited by back EMF, connecting the same motor to 100V would spin it faster before the back EMF effectively cancelled out the input voltage. And, at the RPM that cancels out the 84V, the 100V input still has 16V of headroom. Assuming the batteries and control board can supply the current, there is power available to continue accelerating the motor being supplied with 100V.

Remember that the power required to overcome wind resistance is roughly related to the cube of the wind velocity. When you're already going really fast, going a little bit faster requires vastly more power just when your motor is less able to grab that power due to high back EMF. I like to think this is why KS hard limits my wheel to ~30 MPH even though the same motor and battery configuration in a Begode would reach higher speeds. For whatever reason (inadequate design, conservatism, no guts, crappy parts, lawyers, reputation) KS isn't confident enough that they can deliver the additional power at 35 MPH to keep me upright when there will be gusts, bumps, rises, rider flinches and so on. There are a lot of things working against you when you're going fast.

One other thing... "acceleration" is a result of leaning, but when you hit a bump/pothole you need additional acceleration both to lift the machine (and some of you) up over the bump or out of the pothole, and to speed up the machine to get it back under your feet. When you encountered the hazard, your body kept going forward with very little change in velocity but the wheel suddenly slowed down… the wheel needs to catch up. If you hit the hazard when you're already near the limit (near top speed, going up a steep hill) these sudden peak demands can cause the control electronics to get mad and throw all its toys out of the bed (no-beep cutoff?). Sometimes, just to show you who's the real boss, they get super mad and let the magic smoke out. Both situations rapidly introduce your chin guard to the pavement.

Edited February 9, 2021 by Tawpie

[Rich Sam]

Thanks @Tawpie and @Yellowjacket I just needed a nudge to get my thick head to grasp it!  Very interesting point on EMF I will have to look more into that.  I used to work on electronic boards a looooong time ago and know just enough to be dangerous.   Mostly low to mid voltage stuffs and no motors in my equipment.  I could not noodle how 100v was so much better than 84 volts, but that makes sense based on the laws of power.

[Zopper]

8 hours ago, meepmeepmayer said:

A modern big battery wheel (think V11) is virtually impossible to overlean from anything but too high speed. A V8, you might be able to overpower it at 20kph if you really try and weigh 120kg.

And a weak wheel like Inmotion V5 or Ninebot S2 (not even mentioning S1) can be overleaned by a 60 kg person at practically any speed. When experimenting with it, I had no issue to overbrake it (which is like overlean, but in the opposite direction) even bellow 10 km/h. Which is why I think that new riders should start with those weak wheels, so when they experience the moment when they ask for too much power, it's at a low speed and they won't slam into something at 50+ km/h.

BTW, a great post.

 

Edited February 9, 2021 by Zopper