Tried V8F on a mild slope today

[rcgldr]

It's a small hill in our complex, about 100 feet long and 6 degrees (according to my phone's bubble level app). I weigh about 190 lbs. I rode about a mile first on a 100% charge, then went up and back down the hill. I thought I would have to compensate for the slope, but the V8F tilted forwards when going uphill, and tilted backwards when going downhill, so I didn't have to make any adjustments. Knowing my battery was close to 100% charge, I went slow downhill, and the V8F kept the speed slow, making a mid-hill increase in tilt back and slowing a bit more. I didn't get any beeps (alarm tones), so there was some remaining capacity in the battery, or it's able to handle a 6 degree slope by slowing down and somehow avoiding overcharging the battery. Bottom line, I was impressed by how the V8F handled the hill without me having to make adjustments.

[.....]

They dont slow to stop from overcharging. To slow you against your lean would increase the overcharge dillema. I think the slope doesnt sound too much and it handled it well. I'd imagine the wheel was doing what it simply does, remain level. You may have been feeling like it was doing a lot more than that, but I bet it was a matter of simple perspective. I recall at first, hills felt like the wheel was doing some wierd tilt stuff. In the end, it was simply my brain being confused about how I can keep my feet almost level yet magicall trasport up a hill.  If you run any apps, you can probably review for overcharge and such. My wheels sometimes beep at me when full and going down hill. Luckily, no faceplants yet.

[Eucner]

Riding a same hill first up and then down won't ever overcharge any wheel. You just can't regain more energy than you consumed in climbing. Going first down with a full battery will lead to overcharge and possible shutdown.

The recharge current is related to braking force. These devices can even handle high currents of emergency braking.

[rcgldr]

8 hours ago, ShanesPlanet said:

I think the slope doesn't sound too much and it handled it well. I'd imagine the wheel was doing what it simply does, remain level. You may have been feeling like it was doing a lot more than that, but I bet it was a matter of simple perspective.

Newton's third law for torques applies here

In order for an EUC to accelerate or go uphill, the forwards torque exerted by the frame|motor onto the wheel coexists with a backwards torque exerted by the wheel onto the motor|frame and that backwards torque onto the frame needs to be countered by forwards torque onto the frame due to the rider being leaned forwards, so that the center of mass is ahead of the contact patch. An EUC does this by momentarily decelerating a bit (moves center of mass ahead of contact patch) and tilting forwards, which moves the pedals backwards from under the rider, leaning the rider forwards.

In order for an EUC to decelerate or go downhill, the backwards torque by the frame|motor onto the wheel coexists with a forwards torque exerted by the wheel onto the motor|frame, and that forwards torque onto the frame needs to be countered by a backwards torque onto the frame due to the rider being leaned backwards, so that the center of mass is behind the contact patch. An EUC does this by momentarily accelerating a bit (moves center of mass behind contact patch), and tilting backwards, which moves the pedals forwards from under the rider, leaning the rider backwards.

What I wasn't expecting is for the EUC to make such a smooth transition from level ground onto the hill, and doing the leaning for me.

 

 

Edited September 2, 2021 by rcgldr

[mrelwood]

4 hours ago, Eucner said:

Riding a same hill first up and then down won't ever overcharge any wheel. You just can't regain more energy than you consumed in climbing.

I’m not sure if you’re entirely correct here. Talking about strictly “overcharging”, then yes, but an overvoltage situation (which is what is sometimes referred to as “overcharging“) can surely be achieved, which is what the wheel warns about.

 Riding a shallow hill up slowly uses a modest amount of energy, but when you brake as you go back downhill, the momentary voltage can very easily get above the starting voltage. It still doesn’t have as much energy that the battery has given out, but overvoltage is dangerous because of the actual surface voltage on the battery, not the amount of energy included per se.

3 hours ago, rcgldr said:

An EUC does this by momentarily decelerating a bit (moves center of mass ahead of contact patch) and tilting forwards, which moves the pedals backwards from under the rider, leaning the rider forwards.

The wheel itself doesn’t actively decelerate when the rider starts accelerating. Rider’s ankles are capable of adjusting the CoG front and back.

 But how the wheel reacts when rider starts to accelerate depends on the riding mode settings. With a softer mode the wheel simply allows the wheel to tilt a bit before correcting for the imbalance. A harder mode reacts much sooner, and doesn’t allow the pedals to tilt as much.

It’s about the relation between the input force tilting the wheel forward, and output force to the motor to compensate for the tilt. A 1:1 in/out ratio would always keep the pedals perfectly level. In practice, even the soft mode ratio ramps up, so that the motor doesn’t react much during the first 0.1 degrees of tilt, but as the tilt increases, the motor output gets closer to the 1:1 ratio.

I think the V8F already has some riding mode adjustments. You might want to try them out, they are a great way of personalizing the riding experience!

Edited September 2, 2021 by mrelwood

[supercurio]

4 hours ago, rcgldr said:

In order for an EUC to accelerate or go uphill, the torque from the motor onto the wheel needs to be countered by the torque due to the rider being leaned forwards, so that the center of mass is ahead of the contact patch. An EUC does this by momentarily decelerating a bit (moves center of mass ahead of contact patch) and tilting forwards, which moves the pedals backwards from under the rider, leaning the rider forwards.

In order for an EUC to decelerate or go downhill, the torque from the motor onto the wheel needs to be countered by the torque due to the rider being leaned backwards, so that the center of mass is behind the contact patch. An EUC does this by momentarily accelerating a bit (moves center of mass behind contact patch), and tilting backwards, which moves the pedals forwards from under the rider, leaning the rider backwards.

What I wasn't expecting is for the EUC to make such a smooth transition from level ground onto the hill, and doing the leaning for me.

Yes that's something @Inmotion Global programs so well on their wheels, and unfortunately not a universal characteristic.

On some other wheels instead you feel that the motor control is more akin to a dumb throttle you are controlling by pressing more or less on the pedals, and if more is needed then you have to hang off power pads in order to keep the speed constant up and downhill.

I can compare the V10F which will seemingly continue like nothing happens even up or down rather steep gradients with the Sherman which is the extreme opposite, requiring sometimes an athletic amount of input to do the same despite having roughly 3x the battery and motor power.

[rcgldr]

15 hours ago, mrelwood said:

The wheel itself doesn’t actively decelerate when the rider starts accelerating. Rider’s ankles are capable of adjusting the CoG front and back.

A rider can adjust CoG relative to the pedals, but the EUC has to decelerate or accelerate in order to adjust CoG relative to the tire contact patch at the pavement. In physics terms, internal forces can't adjust the CoG relative to the contact patch, it takes an external force to do that, in this case the force from the ground in reaction to the force from the tire (Newton 3rd law pair of forces).

In order to lean forwards, a rider has to initially push down with their heels, then once leaned, push down with their toes to limit or stop the lean. The EUC responds to this by initially decelerating to move the contact patch behind the CoG (a forwards lean), then switches to acceleration to limit or stop the lean.

In order to lean backwards, a rider has to initially push down with their toes, then once leaned, push down with their heels to limit or stop the lean. The EUC responds to this by initially accelerating to move the contact patch ahead of the CoG (a backwards lean), then switches to deceleration to limit or stop the lean.

You can try this on the ground, facing an open doorway with your hands on both sides the doorway. In order to lean forwards and|or push forwards on the doorway, a person has to push down on their heels, and to lean backwards and|or pull backwards on the doorway, a person has to push down on their toes. Even if a person is not aware of the mechanics, it's how a person leans forwards to cause an EUC to accelerate and leans backwards to cause an EUC to decelerate.

Edited September 3, 2021 by rcgldr

[mrelwood]

7 hours ago, rcgldr said:

A rider can adjust CoG relative to the pedals, but the EUC has to decelerate or accelerate in order to adjust CoG relative to the tire contact patch at the pavement.

Yes, you are of course absolutely correct in your detailed explanation, that is what ends up happening. I was trying to explain this when the amount of power required by a top speed tilt-back was questioned. (My point being that it requires practically an equal amount of power for the rider to start braking because of the initial acceleration required to shift the CoG backwards.)

 My point was that it’s not part of the wheel’s programming, that when it’s time to accelerate, the wheel would first decelerate. But yes, it’s what the rider does to shift the CoG.

[rcgldr]

2 hours ago, mrelwood said:

 My point was that it’s not part of the wheel’s programming, that when it’s time to accelerate, the wheel would first decelerate. But yes, it’s what the rider does to shift the CoG.

On level ground, the initial input from the rider is what shifts the CoG, but in the case of hills and "smart" EUCs like the InMotion EUCs, it is doing all of that for the rider on its own: shifting the CoG and leaning the rider as needed without any perceptible rider input to maintain speed despite encountering an uphill or downhill transition. As posted by supercurio, it seems like nothing happens (he doesn't have to compensate) when encountering hills, unlike his Sherman (where a lot of rider input is required). This is what impressed me with how the V8F handled the hill.

Edited September 3, 2021 by rcgldr

[GPSchile]

This is very interesting. It may be the reason I feel the tilt angle of my V11 changing when engaging some hills. Specially if driving seated, if the slope is long enough (ascending or descending), I notice that the vehicle takes "decisions" on its own regarding the tilt angle. I thought initially it was just my perception resulting from adjusting the vehicle in "Commuting" mode combined with 0% of "pedal sensitivity". But it is more than that since it is not limited for a short term change. If the slope is long enough, this "elevator" effect persists all along and for some meters after arriving to the flat or horizontal level.  About 30m riding on flat surface the vehicle gets back to its normal behavior regarding the tilt control. On short slops this is not obvious. It has to be long enough. 
@rcgldr, your description makes a lot of sense to me and explains this behavior on my V11. Perhaps it is something that has been included by Inmotion on their firmware since some time ago but has never been published accordingly. Inmotion is terrible in communicating their innovations. It is obvious just by checking the names or translations of their features. Insufficient effort is invested in presenting them properly.
Like "normal mode" has evolved into "commuting" mode etc. Or "pedal sensitivity" instead of something more descriptive like "tilt tolerance". V11 had so many innovations but the attention was limited to the suspension system. V12 series includes "balance assist" but its description is vague. Thanks for describing these details and collectively finding what are they about. 

[Tawpie]

23 minutes ago, GPSchile said:

Like "normal mode" has evolved into "commuting" mode etc. Or "pedal sensitivity" instead of something more descriptive like "tilt tolerance". V11 had so many innovations but the attention was limited to the suspension system. V12 series includes "balance assist" but its description is vague.

KS has a setting... "Hand up: unmotivated, damping, and something else". So much clearer don't you think? 

I'm going to have to test to see if KS (or the MTen) do something similar on long hills. I feel like having to "push harder" up a hill is a mental habit I've developed and maybe that's a waste of time. It seems like I need to push harder, but now you got me thinking!

[GPSchile]

during trips to our local mountains I found very useful to adjust the tilt angle accordingly with the app, but it is annoying to stop to do it each time. Something I found can emulate a bit this effect is just to advance the position of my feet on the pedals (just 1 or 2 cm instead of fully centered) when going uphill. To push with lesser effort. Also, this helps me getting my feet closer to the vertical over the contact patch, for better turning or lateral control of the EUC. Going downhill is just the opposite. I center or slightly move backwards my feet to push on the back of the pedal with lesser effort. Just a personal style, no idea how it may work for others. 

[supercurio]

I think the forward tilt @GPSchiledescribes is not specific to a hill, but more generally following an increased sustained power demand, which happens when climbing a hill or riding at a higher speeds in order to fight wind resistance.

By acknowledging the sustained throttle input (and titling forward at the same time) it acts a bit like a cruise control system, which will offset the virtual throttle input until the rider ends the input.

Of course it's not a real cruise control because the wheel still need to balance, but compared to other wheels where input (pressure forward or backwards onto the pedals/body) is rudimentarily linked to output like a dumb throttle would, there's a floating offset which helps making things effortless when needed, and just work as expected on climbs or descents.

And it solves the real problem of having to lean forwards to get the wheel climbing, which is pretty awkward since you're not actually accelerating and you'll literally fall forward on steep climbs without pads since this lean is not compensated by acceleration.

At least, it's my theory for Inmotion's implementation, or how it could be done if not 😉

Edited February 12, 2022 by supercurio

[alcatraz]

How would an euc know you're on a hill/descent? A small diameter wheel needs less input from the rider.

The contact patch moves when you're on an incline. It can explain why it feels different.

[rcgldr]

3 hours ago, alcatraz said:

How would an euc know you're on a hill/descent?

From the amount of torque needed to maintain speed (assuming rider is not currently accelerating or braking). The EUC senses it's on a incline since it senses more torque is required to maintain speed, and senses it's on a decline since it senses it takes less torque or negative torque to maintain speed. For an incline, the motor exerts a forwards torque onto the wheel and a backwards torque onto the frame, which requires the rider lean forwards so that the center of mass is far enough in front of the contact patch to generate a forwards torque that counters the backwards torque exerted on the frame. For a decline, the motor exerts a backwards torque onto the wheel and a forwards torque on the frame, which has to be countered by leaning the rider backwards. The Inmotion EUCs lean the rider automatically without any perceptible rider input for inclines and declines, and in addition, automatically tilt forwards for incline and backwards for decline. 

On my V8F, I have pedal sensitivity set to 100%, so they barely tilt in response to rider inputs on level surfaces, which make the auto tilt forward | backward on incline | decline more noticeable. In my neighborhood, the streets are cambered inwards (using the middle for drainage), and there's a sloped T-intersection at the top of an incline where I can do circles, and as it circles, my V8F will cycle through tilting forward and backwards. 

Supercurior commented that this is essentially a very smart cruise control feature that maintains speed on inclines and declines, balancing the rider as needed and auto-tilting forwards | backwards without any perceptible rider input. I was impressed that Inmotion implemented this feature when I first experienced it.

Edited February 12, 2022 by rcgldr

[rcgldr]

9 hours ago, GPSchile said:

V12 series includes "balance assist" but its description is vague. 

It's a poor translation. It's only used when using the kill switch to tilt the V12 forward or backward onto it's bumpers or the kickstand. With "balance assist" enabled and using the kill switch to tilt the V12 forwards or backwards, V12 will try to keep the tire from rolling forwards or backwards, useful on a 65 lb EUC with a lot of battery weight up high.

[supercurio]

6 hours ago, alcatraz said:

How would an euc know you're on a hill/descent? A small diameter wheel needs less input from the rider.

Probably an EUC doesn't know it's riding a hill or descent versus rolling resistance of riding on gravel, or headwind on flat, or even wind resistance altogether due to speed.

But an algorithm can figure out that in order to maintain a certain power output (regardless of speed even) based on input, it can offset the input like via a moving average of the throttle input over 0.5/1s.

(Though it should be something a bit more suited than a moving average to make sudden changes easier)

That way, instead of asking a rider to remain on its toes the whole climb which is pointless, it could only require to not brake.

[rcgldr]

20 hours ago, supercurio said:

Probably an EUC doesn't know it's riding a hill or descent versus rolling resistance of riding on gravel, or headwind on flat, or even wind resistance altogether due to speed.

I don't know the details of the balancing algorithm, but since the first self balancing unicycle (SBU), one of the required inputs is a magnetometer, used to determine which way is true "up", so there may be something in the inputs and feedback to let an EUC know if it's on an incline or decline.

On level ground, the contact patch is directly under the axle. On an incline | decline, the contact patch is shifted forwards | backwards from directly under the axle, but it probably doesn't matter. What matters is the torque about the axle of the EUC. The motor generates a torque upon the wheel and an opposite torque on the frame. The rider exerts a torque on the frame, and the frame exerts an opposite torque on the rider. If the rider is leaning, then the upwards force from the pedals and the downwards force from gravity at the center of mass generates a torque on the rider. If there is acceleration (forwards or backwards), then there is also a torque exerted onto the rider due to the acceleration acting at the center of mass and the forwards | backwards force from the pedals. The torque exerted by the motor onto the wheel exerts a force onto the ground, and the ground exerts a force onto the tire.

The key factor is when in a balanced state, there is zero net torque on the frame and zero net torque on the rider, regardless of any acceleration, wind, incline, decline, ... .

 

Edited February 13, 2022 by rcgldr

[Tawpie]

Huh. And since such a control system would want to be underdamped (delay reacting to changes in 'which way is down') so it didn't overreact, it might be where pedal dipping out of turns and going over bumpy stuff came from in the early iterations of wheels?