Jump to content

Just unboxed my Gotway MSX 100v, quick initial PRE-RIDE impressions in comments (compared to my inmotion v8)


photorph

Recommended Posts

On 6/4/2019 at 2:43 AM, mrelwood said:

As noted in the brilliant post above by @zeke, switching from acceleration to braking (without tilt-back) requires extra power as well.

Though that's only true if we consider the rider body to be static. Bending the knees and thereby stop countering the torque that pushes the rider forward will instantaneously reduce the current demand and still move the wheel in front of the rider (in particular at higher speed the rider may also fall quickly behind simply from the air resistance). I am pretty sure that one can reduce the current demand and at the exact same time speed up the wheel (without rider) , unless being already at a (very) low speed and at (very) low acceleration. That also works in practice, I have saved a few overlean situations this way. Kind of relevant to realize ;), bending the knees is the invariable life saver.

Edited by Mono
  • Like 2
Link to comment
Share on other sites

4 hours ago, Mono said:

Though that's only true if we consider the rider body to be static. Bending the knees and thereby stop countering the torque that pushes the rider forward will instantaneously reduce the current demand and still move the wheel in front of the rider (in particular at higher speed the rider may also fall quickly behind simply from the air resistance). I am pretty sure that one can reduce the current demand and at the exact same time speed up the wheel (without rider) , unless being already at a (very) low speed and at (very) low acceleration. That also works in practice, I have saved a few overlean situations this way. Kind of relevant to realize ;), bending the knees is the invariable life saver.

I agree, bending the knees rapidly and keeping the pedal pressure at the forefoot should do just that. But that move would do the same during a tilt-back as well, which is why I think there is no difference in power requirement in user-generated and tilt-back generated braking. Other than doing the move after noticing an overlean situation might be more spontaneous than doing it when a tilt-back starts.

  • Upvote 1
Link to comment
Share on other sites

8 hours ago, mrelwood said:

I agree, bending the knees rapidly and keeping the pedal pressure at the forefoot should do just that. But that move would do the same during a tilt-back as well, which is why I think there is no difference in power requirement in user-generated and tilt-back generated braking. Other than doing the move after noticing an overlean situation might be more spontaneous than doing it when a tilt-back starts.

Not really in second you are actively moving/ tilting wheel using bend knee, and static top body,  you save power. In tilt back, the wheel need to generate and reposition feet which is also a change in PID characteristics (not only input) and those who know control theory will see quite a difference.

If somebody is driving like stick than yes, tailback  a braking may require similar power...

Link to comment
Share on other sites

1 hour ago, LucasD said:

In tilt back, the wheel need to generate and reposition feet which is also a change in PID characteristics (not only input) and those who know control theory will see quite a difference.

I do know control theory and I don't see that the setting/adjustment of the neutral angle (AKA tilt-back) would make a difference. Obviously, the wheel doesn't reposition the feet and conversation of energy laws remain intact :D The riders reaction to tilt-back may or may not demand energy from the wheel. Of course I don't know whether the software also changes the controller characteristic under tilt-back. If it does, it seems advisable that it makes the characteristic less power demanding :thumbup:

  • Like 3
Link to comment
Share on other sites

Minimum change would be offset to control,  but most likely it would result as suddenn step function, at least in Kingsong given their brutal tilt-back ;)

To reposition pedal you need to speed up wheel, but it does it with full weight of rider (as it is not conscious), also if somebody has perfect feet control (and flexibility), the tilltback would not effect his COG.

 

Edited by LucasD
Link to comment
Share on other sites

I just bought a used MSX and man is this thing beat up.The panel where the power button and charging port are located is pushed in slightly, and detached from the body panel on one side. It works for now but at some point I'm going to have to fix it before it caves completely. Otherwise the wheel seems pretty strong and powerful.

Link to comment
Share on other sites

  • 5 months later...
On 6/10/2019 at 3:45 PM, Mono said:

Though that's only true if we consider the rider body to be static. Bending the knees and thereby stop countering the torque that pushes the rider forward will instantaneously reduce the current demand and still move the wheel in front of the rider (in particular at higher speed the rider may also fall quickly behind simply from the air resistance). I am pretty sure that one can reduce the current demand and at the exact same time speed up the wheel (without rider) , unless being already at a (very) low speed and at (very) low acceleration. That also works in practice, I have saved a few overlean situations this way. Kind of relevant to realize ;), bending the knees is the invariable life saver.

Indeed, I assumed steady-state behavior to simplify the discussion so it wouldn't be a hundred-page dissertation that puts everybody to sleep. The discussion becomes substantially more complex if we consider that the EUC needs only to accelerate itself and the rider's lower legs, and not the rider's total body, if the rider bends his/her knees momentarily. And yes, reducing the EUC's torque load to increase its speed during a transient is a lifesaver.

The point of my post was to highlight a desire to understand on what basis the warnings are programmed. Are they programmed to maintain a constant torque margin? A constant velocity margin? Or something else? I would love greater transparency so I know how much to trust the warnings.

On 6/11/2019 at 6:19 AM, Mono said:

I do know control theory and I don't see that the setting/adjustment of the neutral angle (AKA tilt-back) would make a difference. Obviously, the wheel doesn't reposition the feet and conversation of energy laws remain intact :D The riders reaction to tilt-back may or may not demand energy from the wheel. Of course I don't know whether the software also changes the controller characteristic under tilt-back. If it does, it seems advisable that it makes the characteristic less power demanding :thumbup:

EUC's are a great, perhaps the best, application of control theory ;) I won't be surprised if somebody out there gets a Ph.D. studying EUC dynamics.

Gradual tilt-back which changes the EUC's "upright" setpoint, as most tilt-backs are, creates such a negligible transient increase in torque demand as to be inconsequential.

  • Like 1
Link to comment
Share on other sites

5 hours ago, zeke said:

The point of my post was to highlight a desire to understand on what basis the warnings are programmed. Are they programmed to maintain a constant torque margin? A constant velocity margin? Or something else? I would love greater transparency so I know how much to trust the warnings.

I can see that this would satisfy some of my curiosity, but I don't see how knowing this would change my actual behavior. To my understanding any warning means that I am getting close to running out of torque, whatever it is based upon. How much torque do I have left? The only way to actually find out is to push over the limit, which I generally try to avoid (though it has happened). I also assume that if, for example, I accelerate too quickly, I may not get any warning at all before I run out of torque. More general, I assume that the wheel does not predict my future riding state (i.e. change of speed or of acceleration) well enough such that I would want to rely upon its predictions.

On my V8 I do get overspeed and overload warnings and I am pretty sure the former is based on a speed measurement and the latter is based on a current measurement (possibly combined with a speed measurement).

Link to comment
Share on other sites

On 6/2/2019 at 7:33 PM, zeke said:

I back off as soon as I feel the KingSong tilt-back. I accelerate until tilt-back begins, back off a bit until it goes away, and repeat a small cycle between activating and deactivating it.

Activating slowly like GotWay means it's hardly noticeable, and once you do notice it takes forever to go away, so the activation and deactivation cycle is slow and large.

Maybe I just need to get used to it though and it'll grow on me ... which won't happen unless they increase the tilt-back speed options.

 

I like your attitude. It lends some confidence that the 3rd alarm is set sufficiently low (leaving enough margins) to be useful.

That said, I think I might have heard it while going to grab food yesterday. I heard a really faint beeping, and it was so quiet that I wasn't sure if it was just my imagination. Maybe the first order of business is increasing the beeper volume; the KS18L's beeper is definitely louder.

With how smooth and stable the MSX feels, it's really easy to get up to those speeds.

 

Thanks for the clarification. People seem to suggest their wheels unexpectedly "cut out," as if they just turned off without provocation, but this makes more sense.

To clarify these discussions, it seems prudent to speak of "torque margins," if you'll allow me to make up a term for it:

image.thumb.png.ff0c3078fe46cd8f8234b800ceda5228.png
 
We can define the steady-state torque margin τM as the difference between the maximum deliverable torque τmax and the steady-state load τload. The maximum possible torque generally decreases with velocity v, while the load torque increases with velocity, as shown above.
 
Some torque margin is always required to support transient torque demands: for example, before you can decelerate, the wheel must first accelerate ahead of your center of gravity so that your center of gravity falls behind the contact point with the ground (causing a net force backward). Also, if you hit a bump the wheel will slow down and must be accelerated to return underneath your center of gravity.
 
The speed at which there is no more torque margin, τM=0, can be described as the wipeout velocity, vwipeout. Upon exceeding this velocity, it is impossible to actuate the wheel to slow you down by zipping ahead of you; it is only capable of falling behind you, which further accelerates your body, and your only option is to faceplant while its only option is to turn off after exceeding its over-lean angle. (Unless you can suddenly increase your air drag, like by opening your jacket... good luck.) It should be noted that, as I have defined it here, this is on flat ground; if a bump is encountered, or if you attempt aggressive acceleration by placing your center of gravity too far ahead of the contact patch, then it is possible to wipe out at a velocity lower than vwipeout. (It is arguable that these gadgets ought to have a boost converter to avoid wipeouts, but that's a debate for another day.)

To avoid wipeouts, the wheels implement audible alarms and tilt-back to communicate to the rider that their torque margins are dangerously low, and the wheel might not support the transient torque demands that it could be subject to. At the alarm speed valarm, some amount of torque margin is still available, τMA: the torque margin when the alarm turns on. Just how much torque margin should be available at the alarm speed to handle these transients under "normal" riding conditions is presumably a matter of debate; hitting larger bumps will require larger torque margins, as will aggressive and heavy riders who attempt to accelerate quickly. Different manufacturers probably have different ideas of how much margins are appropriate. It would be great if we could program our wheels based on how much margin we wanted, but my understanding is that we don't currently have enough information to make such decisions. Maybe if somebody would stick their wheels on a dyno we could at least get a better idea of what these margins are.

 

As the battery discharges, two changes occur: its open-circuit voltage reduces, and its equivalent series resistance (ESR) increases. The former acts to reduce the no-load velocity, while the latter flattens the slope of the torque/velocity curve to produce even less torque still. Both of these cause vwipeout to reduce dramatically. In response to these changes, the alarm speed valarm must reduce in order to keep τMA reasonably large:

image.thumb.png.9a8989bbe7ee35ac0ad575eec627367f.png

It would be interesting to know how the alarm speeds are reduced: Are they reduced in such a manner as to keep the alarm torque margin τMA constant as the battery discharges? Or are they reduced in such a manner as to keep the alarm velocity margin (vwipeoutvalarm) constant? Or something different altogether?

 

Agreed. Tilt-back is simply a way for the wheel to communicate to the rider that the torque margins are low; if you want to test your luck by pressing ahead anyway, it can't stop you. I do however believe that tilt-back is good; it communicates this effectively even when the wind noise drowns out the sound of the audible alarm.

This is an absolutely FANTASTIC post, and should be sticky'd at the top of forums. Understanding these graphs of BLDC motor's torque vs speed is absolutely essential for all riders.

Imagine that the motors we have in EUCs are gas engines. Gas engines have torque that continuously grows with speed, up to the maximum speed determined by max RPM of a piston engine. So, we know that we can just floor the gas pedal and get to max speed, no problem. For EUCs, it's natural (and extremely naive) to think that, well, we can't ride at max speed because the motor won't be able to counteract fluctuations of balance, simply because it can't spin faster; ok, we just give it something like 5kph margin, set some beeps, and then we're totally safe. WRONG. Our motors' torque is approx. linearly falling from standstill towards "theoretical" max riding speed where thrust curve intersects the motion resistance curve (which is predominantly parabolic y~x^2, as at speeds 10kph+ aerodynamic drag is dominating force, over the rolling friction). So, we must pay attention to the quickly decreasing margin of max thrust, as you perfectly explained in your post. I likened this blue curve of thrust vs speed (more precisely, we should look at the difference between blue curve of thrist and green curve of resistance) as walking on thin ice towards the middle of partially frozen river. The only way to be safe is to stay far from the edge.

Edited by Aneta
Link to comment
Share on other sites

2 hours ago, Aneta said:

Gas engines have torque that continuously grows with speed, up to the maximum speed determined by max RPM of a piston engine.

Not wishing to be pedantic but this isn't correct. Gas engines have power that continuously grows with rpm, but the torque falls off way before the max rpm. In fact, the torque can be so poor at high rpm that the power (a multiple of rpm and torque) can drop considerably prior to the max rpm.

In any event, I think I know what you were trying to say, but using a gas engine in the context of your point is probably not ideal. Your comment re BLDC motors losing torque with increased rpm is of course bang-on and should be the focus of every EUC rider :)

 

Link to comment
Share on other sites

1 hour ago, Planemo said:

Not wishing to be pedantic but this isn't correct. Gas engines have power that continuously grows with rpm, but the torque falls off way before the max rpm. In fact, the torque can be so poor at high rpm that the power (a multiple of rpm and torque) can drop considerably prior to the max rpm.

In any event, I think I know what you were trying to say, but using a gas engine in the context of your point is probably not ideal. Your comment re BLDC motors losing torque with increased rpm is of course bang-on and should be the focus of every EUC rider :)

 

Thank you for clarification. Indeed - gas engine:

main-qimg-a69447cd5dc3409a047fcc11d1eb95

BLDC motor:

motorcurve.gif

 

Just to further emphasize the "walking on ever thinner ice" analogy, consider this example in the Motor Simulator:

https://www.ebikes.ca/tools/simulator.html?batt=cust_84_0.2_20&cont=cust_100_200_0.03_V&wheel=17i&frame=cust_1_0.01&hp=0&blue=Lbs&motor=M3540

The no-load speed is about 70kph. It's rather a common ballpark estimate that the max speed of ebikes is typically 80% of the no-load speed, this estimate is often used for EUC's, too, so that would be 56kph max speed. Take several kph for "margin of safety", and we're talking about "50kph wheel". The simulator shows the max speed of 54kph (I doubled both rolling resistance, and drag coefficient, relative to "MTB Upright" configuration, to account for one tire vs. two, and for stand-up body position).

But let's see what the thrust margin at 50kph actually is. Max thrust at 50kph is 19.7kg, while the resistance is 13.7kg (to calculate resistance, take the load power curve (1867W), and divide it by speed). So, the thrust margin is only 6kg! Imagine there's a small bump on the road with only 7% grade (so that the road rises by only 7cm in 1 meter - "nothing"!). Well, to overcome additional 7% grade, the 100kg total weight needs extra 7kg of thrust, but we only have 6! "Ta-da!" - hello, Yer Majesty Faceplant! Gentle bumps like this are not uncommon even on good quality surfaces. It's the speed that totally robbed us of any safety margin.

Compare it to 30kph: max thrust is 47.9kg, resistance is 5.6kg. We have thrust margin of 42.3kg, SEVEN TIMES higher than that at 50kph! At 30kph, the rider can overcome SERIOUS bumps with 46% grade (25 degrees), not that measly 6% (3 degrees) as at 50kph!

Just some juicy food for thought.

  • Like 3
Link to comment
Share on other sites

2 hours ago, Aneta said:

Compare it to 30kph: max thrust is 47.9kg, resistance is 5.6kg. We have thrust margin of 42.3kg, SEVEN TIMES higher than that at 50kph! At 30kph, the rider can overcome SERIOUS bumps with 46% grade (25 degrees), not that measly 6% (3 degrees) as at 50kph!

Exactly. It's good to know how much thrust you have left, so you can gauge how aggressively you should approach the terrain. Better to figure it out on paper, than to figure it out on asphalt.

Although I must disapprove your use of kilograms for thrust, as the kilogram measures mass and not force. The newton is the proper measure. ;)

20 hours ago, Mono said:

I can see that this would satisfy some of my curiosity, but I don't see how knowing this would change my actual behavior. To my understanding any warning means that I am getting close to running out of torque, whatever it is based upon. How much torque do I have left? The only way to actually find out is to push over the limit, which I generally try to avoid (though it has happened). I also assume that if, for example, I accelerate too quickly, I may not get any warning at all before I run out of torque. More general, I assume that the wheel does not predict my future riding state (i.e. change of speed or of acceleration) well enough such that I would want to rely upon its predictions.

In the same paragraph you stated how this knowledge wouldn't affect your behavior, followed by exactly how it should affect your behavior. The alarm is programmed to go off at one arbitrary value of margin, perhaps for a 70kg rider on flat ground expected to hit 5cm bumps with 200kPa tire pressure, perhaps on some other basis entirely, which is not appropriate for all riders and riding conditions. Maybe if you wipe out enough times you can develop a feel for where the limit is, but I'd prefer not to.

Perhaps the best thing would be if the EUC gave us continuous feedback on how much thrust margin it had left, so that with experience we could develop an intuitive feel of how close we are to the limit without having to reach it. It needn't be a precise measure; even just a beeper that beeps faster and faster as you approach zero thrust margin.

Where D is the motor drive duty cycle (ranging from 0% to 100%), the thrust margin is (to first order) proportional to 100%−D, so something as simple as a beep or a chirp that repeats with a period proportional to 100%−D may be sufficient. Better than what we have now, anyway.

Edited by zeke
  • Like 1
Link to comment
Share on other sites

9 minutes ago, zeke said:

Although I must disapprove your use of kilograms for thrust, as the kilogram measures mass and not force. The newton is the proper measure. ;)

To clarify, I intentionally converted newtons to "kilograms of force" in some calculations to make it understandable even for those who don't know physics and for whom the word "newton" is only associated with some ancient guy with a curly wig. 1kg of force is the weight of 1kg mass on Earth, or 9.81 Newtons.

  • Like 2
Link to comment
Share on other sites

21 minutes ago, zeke said:

In the same paragraph you stated how this knowledge wouldn't affect your behavior, followed by exactly how it should affect your behavior. The alarm is programmed to go off at one arbitrary value of margin, perhaps for a 70kg rider on flat ground expected to hit 5cm bumps with 200kPa tire pressure, perhaps on some other basis entirely, which is not appropriate for all riders and riding conditions. Maybe if you wipe out enough times you can develop a feel for where the limit is, but I'd prefer not to.

Perhaps the best thing would be if the EUC gave us continuous feedback on how much thrust margin it had left, so that with experience we could develop an intuitive feel of how close we are to the limit without having to reach it. It needn't be a precise measure; even just a beeper that beeps faster and faster as you approach zero thrust margin.

Where D is the motor drive duty cycle (ranging from 0% to 100%), the thrust margin is (to first order) proportional to 100%−D, so something as simple as a beeper whose frequency is proportional to 100%−D may be sufficient. Better than what we have now, anyway.

I think this is totally doable. If manufacturer measures the max thrust Tmax vs. speed characteristic of the complete motor/controller/battery system (for various battery levels, too), then for any given speed, the max phase current Imax is known. As the rider rides, the current (momentary) phase current Icurr is also known. The expression Tmargin = (Imax - Icurr)/Imax * Tmax will give us the amount of thrust margin. If rider enters the total weight W in the app, then the expression tan(asin(Tmargin/W))*100 will give us the maximum percent grade (of bumps or overall incline) the wheel can still go over at this speed without faceplanting. (in my example above, it's a laughable 6% at 50kph and ludicrous 46% at 30kph) If this value is below some user-set threshold, the wheel will beg, "Please don't faceplant, please don't faceplant!"

  • Like 1
Link to comment
Share on other sites

2 minutes ago, Aneta said:

I think this is totally doable. If manufacturer measures the max thrust Tmax vs. speed characteristic of the complete motor/controller/battery system (for various battery levels, too), then for any given speed, the max phase current Imax is known. As the rider rides, the current (momentary) phase current Icurr is also known. The expression Tmargin = (Imax - Icurr)/Imax * Tmax will give us the amount of thrust margin. If rider enters the total weight W in the app, then the expression tan(asin(Tmargin/W))*100 will give us the maximum percent grade (of bumps or overall incline) the wheel can still go over at this speed without faceplanting. (in my example above, it's a laughable 6% at 50kph and ludicrous 46% at 30kph) If this value is below some user-set threshold, the wheel will beg, "Please don't faceplant, please don't faceplant!"

I don't think it needs to be that sophisticated. And I think setting a single alarm threshold, based on one riding scenario, doesn't communicate enough information to about how close you are to the limit to ensure you don't hit it unwittingly under other riding conditions. (When you place a torque demand, you commit to it by leaning forward. If your demand happened to exceed the EUC's capability, you find out later.) Better to have something that communicates not just that you're close, but how close you are.

The phase current Icurr is Icurr = (DVBATT − VBEMF)/RTOTAL where D is duty cycle, VBATT is the battery's open-circuit voltage at that state of charge, VBEMF is the motor's back-EMF at that speed, and RTOTAL is the sum of the battery ESR and MOSFET and winding resistances RTOTAL = RESR + RMOSFETS + RWINDINGS. The maximum possible current Imax at that speed is Imax = (VBATT − VBEMF)/RTOTAL. Using the first equation to solve for VBEMF and plugging into the second equation, we find Imax − Icurr = (1 − D)VBATT/RTOTAL. Since Imax − Icurr is proportional to the thrust margin Tmargin, then (so long as VBATT/RTOTAL is approximately constant) knowledge of (1 − D) directly tells us Tmargin. Repeated approaches near the limit Tmargin = 0 under various riding scenarios, with appropriate feedback that communicates how close to the limit we are such as a chirp that repeats with period proportional to (1 − D), will allow us to develop an intuition for where it is so we can avoid it instead of just freaking out that we're near it, or worse, accidentally surpassing it.

Now of course, the assumption that VBATT/RTOTAL is constant is a leap. VBATT decreases with state of charge, RMOSFETS + RWINDINGS scales linearly with absolute temperature in Kelvin, and the battery's RESR varies substantially with temperature and state of charge (in ways that I haven't studied). But it wouldn't be too hard to program into the EUC decent guesstimates for these values with a lookup table, or to have it measure them directly. This would make the thrust margin feedback even more accurate and repeatable.

Link to comment
Share on other sites

1 hour ago, zeke said:

I don't think it needs to be that sophisticated. And I think setting a single alarm threshold, based on one riding scenario, doesn't communicate enough information to about how close you are to the limit to ensure you don't hit it unwittingly under other riding conditions. (When you place a torque demand, you commit to it by leaning forward. If your demand happened to exceed the EUC's capability, you find out later.) Better to have something that communicates not just that you're close, but how close you are.

The phase current Icurr is Icurr = (DVBATT − VBEMF)/RTOTAL where D is duty cycle, VBATT is the battery's open-circuit voltage at that state of charge, VBEMF is the motor's back-EMF at that speed, and RTOTAL is the sum of the battery ESR and MOSFET and winding resistances RTOTAL = RESR + RMOSFETS + RWINDINGS. The maximum possible current Imax at that speed is Imax = (VBATT − VBEMF)/RTOTAL. Using the first equation to solve for VBEMF and plugging into the second equation, we find Imax − Icurr = (1 − D)VBATT/RTOTAL. Since Imax − Icurr is proportional to the thrust margin Tmargin, then (so long as VBATT/RTOTAL is approximately constant) knowledge of (1 − D) directly tells us Tmargin. Repeated approaches near the limit Tmargin = 0 under various riding scenarios, with appropriate feedback that communicates how close to the limit we are such as a chirp that repeats with period proportional to (1 − D), will allow us to develop an intuition for where it is so we can avoid it instead of just freaking out that we're near it, or worse, accidentally surpassing it.

Now of course, the assumption that VBATT/RTOTAL is constant is a leap. VBATT decreases with state of charge, RMOSFETS + RWINDINGS scales linearly with absolute temperature in Kelvin, and the battery's RESR varies substantially with temperature and state of charge (in ways that I haven't studied). But it wouldn't be too hard to program into the EUC decent guesstimates for these values with a lookup table, or to have it measure them directly. This would make the thrust margin feedback even more accurate and repeatable.

I agree. (1 - D) in percent is a simple and intuitive metric for estimating the wheel reserves in relative terms. Adding absolute numbers to that would add more detailed info for those more scientifically inclined. For example, "at current speed you can only accelerate at 0.1mph per second, for 5 seconds max; or climb a 1-degree hill; or be hit by a headwind gust of 5mph... but not any combination of these!" - it might prevent faceplants like this:

 

Link to comment
Share on other sites

2 hours ago, zeke said:

The phase current Icurr is Icurr = (DVBATT − VBEMF)/RTOTAL where D is duty cycle, VBATT is the battery's open-circuit voltage at that state of charge, VBEMF is the motor's back-EMF at that speed, and RTOTAL is the sum of the battery ESR and MOSFET and winding resistances RTOTAL = RESR + RMOSFETS + RWINDINGS. The maximum possible current Imax at that speed is Imax = (VBATT − VBEMF)/RTOTAL. Using the first equation to solve for VBEMF and plugging into the second equation, we find Imax − Icurr = (1 − D)VBATT/RTOTAL. Since Imax − Icurr is proportional to the thrust margin Tmargin, then (so long as VBATT/RTOTAL is approximately constant) knowledge of (1 − D) directly tells us Tmargin. Repeated approaches near the limit Tmargin = 0 under various riding scenarios, with appropriate feedback that communicates how close to the limit we are such as a chirp that repeats with period proportional to (1 − D), will allow us to develop an intuition for where it is so we can avoid it instead of just freaking out that we're near it, or worse, accidentally surpassing it.

Now of course, the assumption that VBATT/RTOTAL is constant is a leap. VBATT decreases with state of charge, RMOSFETS + RWINDINGS scales linearly with absolute temperature in Kelvin, and the battery's RESR varies substantially with temperature and state of charge (in ways that I haven't studied). But it wouldn't be too hard to program into the EUC decent guesstimates for these values with a lookup table, or to have it measure them directly. This would make the thrust margin feedback even more accurate and repeatable.

Totally agree. 

I still cannot understand why it is allowed to buy/ride an EUC without this knowledge.

Every buyer should have to sign a contract, that he has understood all this and that his riding style is adapted to it.

@Sebaisn't it irresponsible of you that your WheelLog app doesn't take these calculations into account in the alarm settings? :laughbounce2:

Link to comment
Share on other sites

29 minutes ago, buell47 said:

Totally agree. 

I still cannot understand why it is allowed to buy/ride an EUC without this knowledge.

Every buyer should have to sign a contract, that he has understood all this and that his riding style is adapted to it.

@Sebaisn't it irresponsible of you that your WheelLog app doesn't take these calculations into account in the alarm settings? :laughbounce2:

It appears that I am missing some sort of joke.

Edited by zeke
Link to comment
Share on other sites

5 hours ago, zeke said:

Although I must disapprove your use of kilograms for thrust, as the kilogram measures mass and not force. The newton is the proper measure. ;)

I quite like the use of kilograms for thrust, as it doesn't introduce another level of indirection and confusion :) kg are a perfectly equivalent measure for force, and one that everyone can relate to and understand, because they have had direct experience with it.

Edited by Mono
  • Like 1
Link to comment
Share on other sites

13 hours ago, zeke said:

Perhaps the best thing would be if the EUC gave us continuous feedback on how much thrust margin it had left, so that with experience we could develop an intuitive feel of how close we are to the limit without having to reach it. It needn't be a precise measure; even just a beeper that beeps faster and faster as you approach zero thrust margin.

Incredibly smart. 
KingSong, Gotway, InMotion — ARE YOU LISTENING??? 

Link to comment
Share on other sites

I am so happy that my V8 doesn't beep at all and that it can be put to (almost) silent on any sound. Continuous beeping at difference frequencies to indicate remaining surplus torque is not an option for me, no way.

Edited by Mono
Link to comment
Share on other sites

1 hour ago, Mono said:

I am so happy that my V8 doesn't beep at all and that it can be put to (almost) silent on any sound. Continuous beeping at difference frequencies to indicate remaining surplus torque is not an option for me, no way.

This is a joke, no? I'm not so masochistic as to enjoy wipeouts.

Then again, I hated my V8. Its PID settings felt mushy and unresponsive, even on the most "aggressive" setting. It was under-powered, and even at low speed (where thrust margin should have been highest) encountering small bumps was scary. Sometimes it would deliver, sometimes it wouldn't. I felt like I couldn't trust it, and I was always on the verge of wiping out. I never did, but I came close a couple times in the most unexpected moments. Even the i6 was better, despite its appallingly low top speed. Then I upgraded to KS18L and loved it. Reliable, solid performance. And now I'm on MSX 100V, and I love that even more.

(Knock on wood) I've never overleaned any EUC. But so far, I've just been praying that the warnings are set at a point where I won't hit zero thrust margin on the terrain I happen to be on. I have no way of knowing how true it is other than donning my battle armor and trying it. In dark moments when my sense of self-preservation is low, this is feasible, but definitely not smart.

The speed warnings on KS18L and MSX seem more reasonable on uneven terrain than the V8, but still they provide an unknown safety margin.

~~~

Of course, nobody would want the chirps to occur all the time. At low speed when thrust margin is maximum, this would be annoying and unnecessary. But it is easy to set some threshold, such as when (1 – D) has fallen below 50%, to begin the chirps. In this example, the chirps would begin when the EUC's ability to deliver thrust has reached 50% of its ability at zero speed.

Link to comment
Share on other sites

46 minutes ago, zeke said:

This is a joke, no?

ehm, no. I am perfectly happy with tiltback and I hate beeps. The latter are also easy to miss due to wind noise or otherwise noisy environments. To base riders safety on acoustic beeps is IMHO pretty ridiculous, meaning unsafe. It's BTW also noise pollution which seems to become recognized as a relevant public health concern.

  • Like 2
Link to comment
Share on other sites

4 hours ago, Mono said:

ehm, no. I am perfectly happy with tiltback and I hate beeps. The latter are also easy to miss due to wind noise or otherwise noisy environments. To base riders safety on acoustic beeps is IMHO pretty ridiculous, meaning unsafe. It's BTW also noise pollution which seems to become recognized as a relevant public health concern.

Ah, understood. I mistook you to mean you turn off all warnings.

The reason I don't use tiltback on the MSX is because it's programmed really weird and annoyingly. I really liked the KS18L's tiltback.

At speed, I can barely hear the MSX's beeper. :( A chirp (unlike a beep, which plays a single tone, a chirp is defined to sweep across a large range of frequencies quickly) would be more noticeable. It would also be preferred if the sound was directed upward, instead of to the sides. Ah, the price I pay for a higher performance but poorly designed EUC. :unsure:

  • Like 1
Link to comment
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

×
×
  • Create New...