Jump to content

New rider on giant wheel


svenomous

Recommended Posts

I ended up not riding yesterday (Monday), because I got home from work later than expected, and it was raining!  Wasn't supposed to rain yesterday.  Well, 10% chance was in the forecast I suppose.  It stopped at some point, and there was still some daylight, but the roads were wet, I was settled into full couch potato mode (drool on my face, eyes glued to the TV), and my back was stiff from my morning workout at the gym, so I decided to take the day off from learning.  Today, though, the sun was shining and as soon as I got home I put on all the gear and grabbed the wheel to head out of my house.  Took neighborhood sidewalks to the nearby park & ride lot for some much-needed practice, and later took a tour of neighborhood streets for a few total miles (total km today was about 5, wheel odometer reads 32km now).

My main focus was on improving my mounts.  I was already able to place either foot on its pedal, stand astride, do the triangle thing, and "hop" the other foot up onto its pedal to get started, but:

  • Where the 2nd foot ended up on the platform was more or less random, and often it was just way off the desired position, so an otherwise successful mount was still a failure and I would dismount again to fix this
  • My mount success rate was not good enough (maybe 70%), and it got worse when under pressure (being watched, especially at a crosswalk with a car waiting...oh the embarrassment!); it has to be 99.999%

So, my goal was to mount, roll 5 feet, dismount, alternating from foot to foot, and to concentrate on bringing up the 2nd foot slowly and deliberately, while the other leg was holding the wheel stable "the triangle way," to place it in the right spot.  I did this about 50 times or so, and definitely got better.  I'd say I can now claim a 95% success rate on the mounts, and I can bring up the 2nd foot much more slowly than before (instead of a panicked hop, it's more of a deliberate raising and positioning).  I do look down when I do this, but as soon as the foot touches down on its pedal I look forward and lean into the acceleration.  Part of the technique is to have just the right amount of forward lean already going as the foot comes up, so the wheel is already starting to move.  I don't do "skating" starts, just lean and raise, allowing the wheel to get itself started.  This works great on level and downhill grades, but on uphill I've found that a lot of lean is required to initiate forward momentum and not wobble like crazy.  Maybe I should give a little skate start on uphill grades?  Further improvement on success rate is needed, as well as even more deliberate raising and placing of the second foot.

A secondary area for improvement is the ability to shift my foot while in motion.  The pressure on the foot/leg that has to hold all the weight and lateral pressure while the other foot shifts is significant (especially considering I'm about 95kg at the moment), and also the wheel becomes a little unstable during the attempt, so even after some practice I find that a "shift" is something I can't do reliably or well.  Needs practice.  One problem I'm having is that the wheel makes a loud rubbing noise when it is side-loaded.  While mounting (especially right-pedal-first) and in turns (especially to the left), there's a rrr-rrr-rrr-rrr sound and vibration that I don't like, and that makes me less interested in doing turns or these foot re-position attempts (which also cause the rubbing sound).  I wonder if this is normal, or if something is not aligned correctly in this wheel. 

Another challenge I wanted to get a handle on is deciding on correct foot positioning for comfort and efficiency.  After reading some more posts, I've realized my feet were placed too far rearward.  I did have them forward of center (toes overhanging more than the heels), but not enough.  I've added about an inch of forward overhang and the foot strain/pain is much reduced.  I'll keep experimenting with this until I find just the right position that gives me maximum comfort while also allowing for easy deceleration.  I've also just read tonight in another thread about "duck-footing," as in angling feet outward a bit, and that's something I want to try out.

My turns are getting better.  Today I used a storm drain in the middle of the parking lot as a center reference point for figure-8's--making a point of rolling over the drain each time to add a little stability challenge--and I was able to make them the tightest and most consistent yet.  OTOH, when riding down a sidewalk or bicycle lane (yes, I've ventured onto bike lanes in my neighborhood!) in a straight line, somehow my turning confidence erodes, and when I then have to make a 90 degree turn onto another street, the turn is a mess.  I don't get it.  While in the parking lot doing nothing but turns, all is fine and almost subconscious, but when out cruising the turns suddenly become an over-aware wobbly disaster.  I've also had moments of sudden loss of certainty and coordination, usually after a moment of distraction, and suddenly the wheel felt like it wanted to go off in one direction without my body making the subconscious correction needed.  Sometimes I'm concentrating too hard on the body stance and movements, when I should just let the cerebellum form its natural pathways.

Stance...I tend to crouch and lean forward during turns, and even when going straight I have to keep telling myself to straighten up, push out the chest, and not slouch over my knees so much.

As for traffic, I was out during the tail end of the evening commute, and I live in a suburban enclave for tech workers and their families, so a steady stream of minivans and SUVs was on the neighborhood streets.  In most cases when I had to cross a street from a sidewalk, or via a bike lane, if there was any conflicting or turning traffic I would stop dead and wave them on, as I'm not yet confident with big head movements to check for traffic over my shoulder, nor with my ability to safely and quickly make the S-turn transitions from sidewalk to the crosswalk lip, across the crosswalk, up the other sidewalk lip, and back onto the sidewalk.  Having to awkwardly dismount in the middle of an intersection because I lost my balance, while a car is waiting to proceed, would make me sink through a hole of mortification, never to be seen again.  Still, this was the most traffic I've dealt with, and it wasn't so bad.  My directional control is mostly good when going in a straight line, so I was able to maintain a very predictable path for pedestrians and cars not to feel freaked out.  Still, one minivan almost hit the median regardless, while trying to stay as far away as possible from the helmeted alien on the death glider.

OK, that's all I have today, I think.  I've already decided (I think I mentioned this in another post) that after I get back from my hiking trip I will do more local practice, and in 2 weekends I will go onto one of the heavily trafficked local bike/pedestrian trails, in broad daylight, among joggers, walkers, parents with baby strollers, dog-walkers, bicyclists, skateboarders, children, the occasional family of ducks or geese, etc.  Slow and steady, dismount when it feels prudent to do so, and go a few miles up the trail and then back down to the park where my car will be waiting.  Building confidence.  If that goes well I will probably do a work commute dry run that same weekend.  It's 8.3 miles each way, on a road shoulder (not quite a bike lane, but close), then a bike lane, then dedicated bike/pedestrian trails with some street crossings.  No actual "in traffic" riding as you see in city riding videos, but the road shoulder and bike lane is heavily trafficked on weekdays (these are commute arterials), and there are plenty of side street and driveway crossings.  I do this route 2-3 times a week on a bike, and need to build confidence before attempting it on a circus clown unicycle, but it's my determined goal to do some of my future work commutes on the wheel.

  • Like 1
  • Upvote 1
Link to comment
Share on other sites

On 5/1/2019 at 8:58 AM, svenomous said:

area for improvement is the ability to shift my foot while in motion. 

I think that instead of rehearsing getting the foot placed perfectly every time, I would concentrate on this. My 2nd feet rarely lands at a good position, so I move it around while in motion. And when facing a steep long downhill, I do the manoeuvres to get a bit backwards on the pedals. Or whatever each situation asks for.

I think the skating start / kickstart / pushed start is the way to go, for several reasons:

- Starting from behind even the smallest obstacle is a huge strain for the electronics. The modern wheels don’t really burn the Mosfets anymore, but also your balance might be disrupted if the wheel struggles a bit to start moving.

- You have much more time to get your second foot aboard (and in position), as you learn to ride one-legged for the second or two.

- You won’t accidentally step on a wheel that’s turned off. Really! I’ve seen this happen multiple times! Having the wheel under just one leg doesn’t clearly tell you if the wheel is powered on or not. If you jump on a wheel that’s powered off, you have nothing to jump off of as the wheel will just roll out from under you.

But if you do the kickstart, you will know right away if the wheel is not balancing, and you will not put your another foor on top of the wheel, so it will be a non-event instead of a possibly dangerous fall.

 

Edited by mrelwood
Link to comment
Share on other sites

I'm back from my hiking trip, and after a couple of days of work hell I've had some time to play with the wheel again!

@mrelwood, thanks for the advice.  I agree that it's almost impossible to place the 2nd foot right during mounting, even though I do think there is value in being able to bring up the 2nd foot "deliberately," instead of in a rushed near-panicky way :).  As for kick starts, let me try them (haven't done any yet).  I get your point that when on rougher terrain, or uphill, or if in any way there is resistance to forward movement, getting a little momentum going is probably a good idea.  Yet, I'm also proud that I can (mostly) just step on and get moving without paddling my way forward first.  Perfection probably lies in between somewhere, with a single brisk kick-off and a controlled and deliberate lifting of the second foot onto its pedal.  I should have an opportunity for more practice tomorrow evening, as the weather here in the Seattle area has been beautiful this week!

My first update since last week is that I went ahead and purchased an Apple Watch.  I'm in iOS ecosystem and didn't relish the idea of on-boarding to one of the cheaper Wear OS or Android based solutions available, and having to swap phones and/or watches just for riding purposes, so I gave in to the Dark Side and set aside my beloved chronograph for the Apple Watch I had previously scoffed at.  It arrived right before my hiking trip and I got it set up and customized, and I even got a hiking companion app installed that turned out to be quite useful on the hikes in Utah: WorkOutDoors.  It can store offline map tiles as well as gpx tracks (which I got out of AllTrails for the hikes we planned on doing), right on the watch, and it was very useful to be able to raise my wrist and glance at a tiny little map that showed whether we were still on the hiking trail (and sometimes we weren't!), and which way the trail was, or which way it was about to bend at the Y junction, etc.  The trails we were on weren't always well marked.  That, plus of course tracking distance and calories and heart rate and whatnot.  Anyway, back to how this applies to the wheel: DarknessBot integration works very well so far, all I have to do is start the app on the phone and let it connect to the wheel, then put the phone in my pocket and start the app on the watch.  The app gives me the basic stats I want, like speed and battery level and temperature.  In addition I can also use WorkOutDoors (I tell it I'm going "skating" when I get on the wheel), and it tracks my movements and speeds accurately while I'm out there training.  Btw the watch says today my max speed was 15.9mph, while DarknessBot claims it was 17.0mph, so clearly the well-known KS speed/distance discrepancy applies, as I trust the watch's GPS-based numbers far more than KS's.  I wear Flexmeter double-sided wrist guards and wondered how I would mount the watch, but that turned out to be pretty easy:

20190508-015518455-i-OS.jpg

The FlexMeters have a Lycra/Spandex sheath between the hard foam/plastic protective plates, and I simply cut a "window" in this material on the left one.  Then all I had to do was put a sport band on the watch (a flexible velcro-fastened band) and place the watch on the side of my wrist instead of on top.  The FlexMeter slips on over this, and the watch just sits there in its "window," easily raised for viewing during a ride.

Training-wise, I'm getting more confident and have raised the alarm/tiltback speed to 20mph.  I've set them both to the same speed as this actually starts a subtle tiltback before the alarm sounds, which I prefer as a more subliminal message than the voice or the beeps.  I've also turned off the voice, finally, since as much as I've enjoyed my conversations with Mr. KingSong, I'm a bit embarrassed when the wheel loudly proclaims "please de-cel-er-ate" just as I'm passing a pedestrian on the street.  I'd rather have a mysterious set of beeps.  My turns to the right are still better than my turns to the left, don't know why.  I can do pretty tight circles and figure-8's, and generally have full control of my direction of travel, but every once in a while my right leg will straighten just a little and the wheel will veer right, requiring a correction, and I don't know why my body is doing that.  Still, no crashes since that first one before I started wearing the gear.  I even took the EUC Bodyguard off today to see what it would feel like to ride without it stretched over the chassis.

Sometimes at speed I get a bit of a wobble, basically a high-frequency oscillation, which I can dampen by a slight knee-bend or by pushing one knee or the other slightly against the chassis, or by leaning into a slight turn, or by decelerating a bit.  From everything I've read these wobbles will resolve themselves with time, as I become more comfortable and relaxed.  I'm also continuing to try different foot positions (forward, back, duck-footed, outboard, pressed all the way inboard, etc.), to see how that affects comfort as well as wobble.  No conclusive results except that slightly duck-footed is a little more comfortable on my knees.  The soles of my feet still hurt after just a few minutes, which surely is due to lack of relaxed stance and due to my feet trying to "grab" the pedals for dear life.  It's better than it was, but I have to stop after about 10-15 minutes of riding, to let the tingles and pain subside, before remounting.  I think I'm also ready to play with calibration, as currently I have the wheel set at a perfect level pitch.  A couple degrees forward seems to be what many prefer, and I plan to try that tomorrow, and see how that affects comfort as well.

Today I did almost 8 miles, mixed between sidewalks, bike lanes, even some street crossings and transitions including roundabouts.  I feel comfortable looking behind me for traffic, signaling turns (although sometimes the act of signaling induces a wobble as my CG shifts), and going over storm drains and up and down sidewalk lips.  I even did a short stretch on a neighborhood dirt trail today, to see how I would do with more rough terrain.  It was a lot of fun, but the trail is very narrow and there was an old lady walking veeeryyyyy slooooowwwllyyyyy ahead of me whom I didn't have the courage to pass, so ultimately I turned back and returned to the paved roads.  Gotta find more dirt trails, though, as I could tell that going over pebbles and tree roots and gravel and uneven ground was really helping me with learning subliminal control and balance.

I believe I'm still on track to venture onto a heavily populated shared-use paved trail on Saturday, one of the major local recreational trails that has bikes, pedestrians, and everything in between on it on any given day.  If that goes well, my intent is still to try a commute dry run on Sunday, where I'd have to travel on the shoulder of a "real road" with "real traffic," as well as on the bike lane of another "real road," and finally on some dedicated shared-use trails (with some road crossings) to my workplace.  I'm anxious about this, but I won't build any more confidence for it than I already have, and the only way to know I can do this is to, well, do it.

  • Like 1
  • Upvote 2
Link to comment
Share on other sites

Btw the dirt trail I was on today was wide and smooth at first, and then got increasingly narrow and uneven.  By the time it got really narrow, with bushes encroaching and the river embankment right next to me, I finally decided to dismount and walk the wheel for the last several hundred feet to get back to the paved trail.  This is the photo I took as I lost my nerve and dismounted.

20190512-203014145-i-OS.jpg

Link to comment
Share on other sites

10 hours ago, svenomous said:

I have to stand straight, with knees almost locked, or I get a speed wobble.

As you seem to be more on the analytical side, try this: step on your wheel without looking at your feet and foot positioning at all. Adjust your stance to something comfortable without looking, either during riding or by holding on to a pole/wall. Ride 100m without looking. Then look down (while still riding). You might be surprised how "bad" the right stance looks, soles overhanging, "too far" in the front or back, etc. But relaxed = good, so this stance is the right one.

10 hours ago, svenomous said:

So, based on multiple rides I've concluded that the wheel over-reports both speed and distance by about 18%.  Does that sound about right?  I wonder why that is...I get that the calculations are based on wheel rotation and not GPS, but bicycle speed sensors also work on that principle, and they're far more accurate than this.  According to the wheel, my total mileage is now 96.7 miles, but if I apply a correction factor it's probably closer to 82 miles).

Wheel distance and GPS distance are not the same. Both numbers are "correct".

Look at a GPS track on a map and how inaccurate it is - a very rough polygonal line, sometimes besides the road instead of on it, cutting corners, long measurement intervals, etc. But that measured distance is what you find on maps and signs as "the distance" to somewhere.

Meanwhile, in real life you may be weaving and doing all sorts of tiny corrections and deviations from going a straight line. That's what the wheel measures (by counting the rotations, and probably with not too accurate of a result as the precise outer tire diameter can vary by a few mm - but not 18% inaccurate). Arguably, this is the more correct number (it's a finer measurement than GPS), and it's going to be greater. Get a real measuring wheel and drag it precisely in your track to see how good this number is. Your 18% difference is plausible.

Unfortunately, that's not the number that corresponds to distances given on maps and signs.

I wouldn't be surprised if wheel odometers (and car odometers!) manually adjust their result so it matches the externally given distances that people expect, precisely so this question doesn't get asked and people think their vehicles give them wrong numbers.

10 hours ago, svenomous said:

<battery stuff>

There are lots of inaccuracies with battery level:

  • Voltage is what the wheel actually sees. Voltage measurements are very inaccurate. Something like +/- 1V reported for some wheels. That alone can be a 5% difference.
  • The voltage drop under load makes any precise voltage measurement meaningless anyways. You can only compare voltages under the same conditions (like: standing still with no rider, or with a certain rider at x mph).
    In the end, the voltage of your battery at each moment is just what it is.
  • Most (all?) battery levels are just disguised voltages, not actual battery levels (aka how much energy is left) because these are really complicated (you have to take into account the discharge curve for your cells under the specific conditions). Voltage is what really counts for how the wheel behaves and what it can and can't do, so this isn't really a bad thing, but it's a thing.
    That means the initial "percent" (voltage in disguise) get you further than the final ones, as the voltage drops a lot faster towards the end. 100% to 90% gets you far, 25% to 15% gets you much less.
    And different apps (and chargers?) may compute different things. Wheellog tried to estimate the actual energy for some time before going back to voltages, and I believe the Inmotion V10 app might estimate actual energy left (not sure).
  • Wheels have different 100% and 0% voltages (numbers per cell, multiply by 20 for 84V wheels to get the actual voltage). For example:
    * Gotways go from 3.3V (0%) to 4.1125V (100%), but you can charge "over 100%" by keeping the charger in and getting the full 4.2V the cells can do. That's an extra 10% in energy and therefore range!
    * The 18XL goes from 3.0V (0%) to 4.2V (100%) instead.
    Apps have to consider that and different apps (and app versions) may give different results.
    Your charger doesn't know which wheel it charges. Is 4V per cell ~86% (Gotway) or 83% (18XL)? Is 3.5V 24% (GW) or 42% (KS)? Same voltage, but your charger doesn't know which percentage it should be. 81.5V corresponds to ~95% for GW and ~90% for your 18XL. (Actually, the interpolation between 0% voltage and 100% voltage is not just linear, so these numbers aren't even exact.)
  • I'm sure there's more. Reality isn't theory, and God knows what dumb stuff the manufacturers do.

So different numbers appearing are no surprise. It's just how it is:)

I found I got a very good intuition for how far and fast I can go with my battery, getting home on the last beeps and with low battery tiltback (but getting home:eff034a94a:). That intuition will also include all other factors playing a role. You'll get yours, too. Don't worry too much about some crazy numbers. EUCs are very empirical (for now, maybe forever).

Edited by meepmeepmayer
  • Upvote 2
Link to comment
Share on other sites

@meepmeepmayer, thanks:

Trying foot position without looking down: will do the experiment.  And yes, I am very much on the analytical side :).

Speed/distance: I don't disagree with much of what you said, but I know for a fact that the distance from my house to my workplace is 8.3 miles.  More than one bicycle computer, properly set for wheel diameter, has said so, and more than one expensive GPS-based bicycle computer has agreed with that number, plus or minus a few hundredths.  The GPS app I used to come up with 16.7 miles for the round trip a few hours is unproven, but 16.7/2 is 8.35, which is pretty close to what I've always seen from hundreds of bicycle trips.  I wasn't S-turning all over the place, so I'd expect variance between "map" distance and actual "tire" distance to be very small, despite the small corrections that you rightly point out.  1% at most.  On Saturday I did a 63 mile training bike ride, which ridewithgps.com's route map said was 63 miles, and when I was done the bike computer agreed that I had traveled 63.1 miles, despite not traveling in the exact centerlines of the mapped roads and minor deviations to enter parking lots for breaks and lunch, etc.  There is no way the wheel traveled 19.9 miles during my 2nd ride today, nearly 10 miles each way.  That's grossly incorrect.  Also, while I wasn't exactly keen to keep looking at displays while on this little adventure, given my current skill level, I noted that wheel-reported speed (per the DarknessBot app) was quite a lot higher than my GPS speed (per the GPS app I mentioned) at any given moment in time.  To say simply that "wheel speed" is different and equally valid than actual speed over the ground makes no sense to me.  There is gross error in play here, and KS should address it, as if nothing else the wheel's master odometer says it has many more miles on it than it actually does.

Battery stuff: I understand a little bit about batteries, and their voltages at rest and under load, and why it is difficult to characterize charge level based on voltage.  That said: 81.5V corresponds to 90% per KS18XL definition, but shouldn't a charger do a full CC/CV cycle with 81.5 as the cutoff, vs. just bulk charging to 81.5 and then stopping?  If a charger took a pack up to 84V and shut down the moment the pack reached that voltage, it would not have charged the pack to 100% since it didn't taper the amperage toward zero.  So, should a charger trying to achieve 90% take the pack to 81.5V and then taper the amperage, or should it just stop the charge like the eWheels one does?  When I power on the fully charged wheel (the voltage having had time to settle to its resting level after the charge), the KS app says it is at 100%.  When I power on the 90%-charged wheel, OTOH, the KS app thinks it's in the low 70's.  Seems to me that the eWheels charger actually under-charges at its 90% setting.

Link to comment
Share on other sites

Very much agreed Kingsong should get their KiloMeters right, not only is adding in the pants ridiculous and amateurish, but for countries who have strick speed limits it handicaps your wheel 😑

Link to comment
Share on other sites

Ok, maybe your Kingsongs do have a problem. On my ACM the difference is smaller (<10%) and, looking at the seriously crappy GPS track on my phone that produces the GPS distance, it's not hard to see why.

6 hours ago, svenomous said:

That said: 81.5V corresponds to 90% per KS18XL definition, but shouldn't a charger do a full CC/CV cycle with 81.5 as the cutoff, vs. just bulk charging to 81.5 and then stopping?

I don't know. I don't know much about how this electricity stuff works.

I was always under the impression that the constant current - constant voltage charging behavior happens automatically when you push electricity into something as the voltage difference decreases.

So the current simply drops linearly from some starting value to 0 proportional to the voltage difference going from some value to 0 (that's just the definition of resistance, which is assumed to be constant for the wheel electronics). The constant current phase is artificial and only happens because the charger limits its max current, otherwise as soon as you connect a huge and unconstrained voltage difference, the charger, cabling and batteries would melt/explode instantly (all your hopes for a constant resistance gone:efef77eaf5:). And the constant voltage phase is simply the asymptotical end behavior as the battery gets filled to 100%.

Then, it would make no sense to have to have a (almost) constant voltage phase when you're still in the domain where the max current setting controls the behavior of the charging curve or where you have a falling current but the voltage isn't constant quite yet. So an instant cutoff at 81.5V makes sense.

Maybe what I think how charging works is completely wrong, and the CC-CV thing is intentionally designed (and could be done otherwise). Then I have absolutely no idea what the answer to your question is. @Chriull Help:efef6b27e5::efef62fc70:

Link to comment
Share on other sites

To clarify the behavior I see from the eWheels charger set to 90%: it charges at constant current (5A in this case) until the voltage reaches 81.5V, and then stops completely.  For 100% it does what you'd expect, where it brings voltage up to about 84V and then tapers amps down toward zero while keeping the voltage steady (constant voltage mode), and when amperage approaches zero the charger stops the charge.  So my question to @Chriull would be whether 90% SOC is achieved the way the eWheels charger is doing it.

Link to comment
Share on other sites

1 hour ago, svenomous said:

So my question to @Chriull would be whether 90% SOC is achieved the way the eWheels charger is doing it.

According to https://batteryuniversity.com/learn/article/charging_lithium_ion_batteries not exactly. In table 2 the soc for cut off and full saturation charges are listed.

Saturation charge (cv phase) is only needed to pump carefully the last ~15% into the cells without destroying them by too high voltages.

And for the balancing circuits, which mostly just get active within the cv phase.

Edited by Chriull
  • Upvote 1
Link to comment
Share on other sites

Based on reading at the link you provided, @Chriull, it looks like CC charging to the cutoff voltage gets a cell/battery up to about 70%, and then CV taper-charging gets it the rest of the way.  So, if the eWheels charger were to take voltage to 84 volts or so, and were to then cut off without entering the CV stage, it would have managed 70% SOC.  The only way to achieve 80% or 90% seems to be to either bulk charge to 84V as if intending to top off the pack, start CV taper, and then terminate the charge early, or (which makes more sense to me) to pick some lower voltage cutoff (like 81.5V), do CC to that value followed by a CV taper down to negligible current.  The eWheels charger's observed behavior, in my opinion, means that at 90% setting it is getting the battery pack nowhere near 90% charge.  In fact it's not even getting it to 70%.  @Jason McNeil, any comment on this claim I'm making?

Link to comment
Share on other sites

1 hour ago, svenomous said:

Based on reading at the link you provided, @Chriull, it looks like CC charging to the cutoff voltage gets a cell/battery up to about 70%, and then CV taper-charging gets it the rest of the way.

I'd see cc up to 4.2V (84V) getting 85% SoC.

81.5V ~4.1V per cell should lead to about 80%.

From the values of table 2.

Ps.: Since this was mentioned imo so e posts before - the shown charge % of the wheels do just correspont very very weakly to the SoC of the cells.

Link to comment
Share on other sites

Sorry, I should've read more of the (long) article, and I only fixated on an early note that said that a high-amp charge to cutoff voltage would get a battery up to 70%. This table (that Chriull mentioned) is the real gold for this particular question/discussion, and of course it aligns with what Chriull has said:

Li-Ion-SOC.jpg

So to achieve 90% on an 84V pack, a charger could either do a full saturation charge (CC + CV) to 82V, or it could do a CC bulk charge up to 84V, and then do CV for just a little bit longer to get up to 90% (but this would be less certain as the initial amperage starting from would be a factor).  So, the eWheels charger should be doing a CC/CV cycle to 82V, not a CC-only cycle to 81.5V, to achieve 90%.

So what I'm getting with 90% setting on this charger is about 80% SOC, maybe a bit less when set to 5A.  I could probably get closer to 80% if I have amps set to a lower value, which is something I'll try.  Good to know.

Link to comment
Share on other sites

7 minutes ago, svenomous said:

So what I'm getting with 90% setting on this charger is about 80% SOC, maybe a bit less when set to 5A.  I could probably get closer to 80% if I have amps set to a lower value, which is something I'll try.  Good to know

I did not look into the numbers, but maybe the annonced 90% from the charger correspond with the charge % shown from the wheel and not with the SoC of the cells?

Link to comment
Share on other sites

I think yes, it would be optimal for the charger to initiate the CV stage at the set 90%. It seems the option to charge only to ”90%” is just a very crude addition to the charger, with minimal amount of added components or change to the existing charger circuitry. I wouldn’t be at all surprised if it’s just a requested feature that was added very quickly without thinking it through very precisely.

The KS odometer overshoot is a known issue, and yes, KingSong should fix it. Good luck waiting for that to happen though...

The EUC world is not the best place for very analytical or pedantic people to be in. It is what it is.

  • Like 1
  • Upvote 1
Link to comment
Share on other sites

Yesterday's rides greatly raised my confidence and comfort, perhaps to unhealthy levels.  Since there's rain coming to my area and I likely won't get to do a "for real" work commute until at least Thursday, if at all this week (as I don't want to ride in rain--yet--on such a poorly waterproofed machine), I went out for one last ride in fairly nice weather this evening.  This was a run to the nearby shopping center to get cash at my bank's ATM.  An excuse.  I learned one new thing, though!  Kind of obvious, I know, so please not too many "duh" type comments, but mounting is much less uncertain and error-prone if one doesn't actually look down at the pedals while doing it.  My fixation on foot positioning has caused me to always look down while mounting, to try to get my second foot in the right spot, but if I just stand astride with one foot in position, and look forward heroically as I mount and begin moving, I can bring the other foot up and more or less land it on its pedal, and more importantly get moving forward without much fuss, pretty much 100% of the time.  Then I can shift and shuffle the foot around if it doesn't feel like it's positioned right.  Obvious, sure, but a revelation to me.  Did some practice in a parking lot on the way home, and I think I'll have a much better time next time I decide to go out into traffic and have to mount to cross an intersection while being observed.

@mrelwood, I'll say that there is indeed room for very analytical people in the EUC world, because here I am and if I want to analyze something to death I will.  It makes me happy to feel like I understand stuff, and what's the harm in asking for things to be better than they are, even if there's little chance of it happening?  I see frequent threads about things people would like to see created, updated, improved, re-thought, analyzed, etc.  Otherwise what's the point of a community where we social simians can gab at each other?  As for pedantic, I'll ignore that as it feels mildly insulting.  

Now back to the battery charger thing, to continue my over-analysis.  I have some data based on the above discussion about battery state of charge vs. voltage with and without CV tapering.  The eWheels charger may not do a very good job of achieving 90% when set to 5A output, but in addition to the percentage knob it also has an amperage knob to play with, and lower CC amperage allows coming closer to a "full" charge for a particular voltage target, so play with it I did.  On two separate charge cycles where I first allowed it to charge the wheel to "90%" at 5A, I next proceeded to do an additional "top-off" at a lower amperage setting.  The first experiment was at 2A, same as the stock charger, and the result was that when the charger stopped its CC bulk charge, my meter had registered an additional 91Wh being pumped into the battery pack (6% of the pack's rated capacity), and after the voltage had settled DarknessBot declared the battery to be at 77%, while the KingSong app declared it to be at 82%.  After today's ride I again charged to "90%" using 5A, waited for the voltage to settle a bit, and re-engaged the charger at its lowest amperage setting of 1A.  This time it pumped an additional 140Wh into the pack (9% of the pack's rated capacity), and the battery percentage per the two apps after a period of time for voltage to settle was 81%/85%.

So, while the charger doesn't do what I would expect it to, for future reference when I want to get the wheel to about about 80% SOC I will use 90%/2A setting, and when I want to get to 85% SOC I will use 90%/1A setting.  For 100% of course I'll just use 100%/5A since at "100%" the charger performs a proper CV tapered top-off regardless of amperage setting.  I take comfort from having some numbers on which to base my future charging actions, vs. just blindly trusting that the "90%" setting on the charger does what it claims, and then possibly wondering why I'm not getting as much range as I expected.  So, this has made me happy, and it has added a little bit of information to the community that might benefit someone else with an eWheels charger.  At some point I'll play with the "80%" setting on the charger, but unless it behaves drastically different from the 90% setting (I'm guessing it just sets an even lower CC target and shuts off without a CV taper), I won't report back on it, especially since it might make me seem...pedantic.

  • Like 3
Link to comment
Share on other sites

11 minutes ago, svenomous said:

So, while the charger doesn't do what I would expect it to, for future reference when I want to get the wheel to about about 80% SOC I will use 90%/2A setting, and when I want to get to 85% SOC I will use 90%/1A setting.

S(tate)o(f)C(harge) is normally used for the "real" charge of the cells. But from he wording it of course is also what the apps "try" to show us...

Wheellog (as imho also the GW app?) shows 100% above 4,11V cell voltage and 0% for below 3,3V and inbetween a linear approximation. (charge % = (V-52,9)/13 ... since all GW report internally just 67.2V the 52.9V have to be adjust for an 84V wheel to 52.9/76.2*84)

So 85% shown by the app mean some 4.0V per cell. That'll mean according to the abovementioned table 80% SoC of the cell with full saturated charge.

Maybe a bit more in regard to SoC after charging with CutOff vs. Saturation charge as mentioned on Battery University. While the CC phase the cell voltage is pushed up (as you correctly lowered the current while the CC phase to "get more charge" into the cell), but the cell voltage will settle again after this - if one lets it settle long enough imho it will reach something around the voltage for this SoC like one has in the table after a saturation charge. (Imo it will also settle just a little bit after a saturation charge?)

So after a 4.1V "cut-off" charge leading to 80% SoC the voltage should "drop" to 4.0V as the 4.0V saturation charge leads to 80%, too.

By this, multiple cut off charges one after another lead to higher SoC, since the voltage settles inbetween and the more often this is repeated the more one approaches a saturation charge.

Link to comment
Share on other sites

 

Quote

As for pedantic, I'll ignore that as it feels mildly insulting.

I’m terribly sorry! The wording I used must have a very different tone in my native language, as I absolutely meant nothing even slightly negative with it. Perhaps ”meticulous” would be a better word for the skill to examine and analyze things in detailed precision?

What I meant was that there are a lot of things with EUCs that are indeed imperfect, many even badly designed or executed, and most of them without proper reason, some even stupid or plain oversights. If one wishes to purchase a thoroughly designed and tested high-quality product with few or no issues, an EUC is not where one finds that. That’s all.

There’s absolutely nothing wrong with questioning, wondering or pondering about the things that are not properly done and should be improved on. It is one of the core fruits of this forum in my mind. And many of those things have been improved because they have been brought up here! 

As long as one doesn’t let them devalue the hobby in one’s mind, or the enjoyment it can bring.

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

@Chriull, thanks, although my head is spinning a bit.  To not over-analyze this, I think that for normal commute and "having fun" charging I will use the 90%/1A setting on the charger, which will give me something in the 80%(ish) range, for medium-term storage (up to a month) I will use the 80%/5A setting, which will probably result in something like 60-70% charge level, and for long-term storage (not that this will ever happen) I will run down the battery to about 40-50% (let's say about 72V resting voltage) and not charge.  This is good enough for me...for now.

@mrelwood, no worries, I was only "mildly piqued" by the word.  Pedantic, to me, is a trait that involves being preachy and too prone to over-explaining things to people.  Of course the reason this piques me is that it's at least partially true :).  Btw I'm having all kinds of fun with this hobby, even if I like to over-think some aspects of it.

  • Like 3
Link to comment
Share on other sites

It's raining here this week.  I'm very sad.  Little riding opportunity.  No "real" wheel commute to work yet.  Next week?

Today I saw a window of opportunity, with streets semi-dry and only light drizzle falling.  The opportunity was to try out some rain gear!  Rain-proof overpants and jacket from my bike gear, and also a new set of knee pads that can be strapped on over pants (instead of slipping them onto my bare legs).  Also, the Bell motorcycle helmet with the full visor, instead of the lighter mountain bike helmet without the visor.  I finally had to figure out how to solve something I've been wondering about: while wearing the Flexmeter wrist guards, what if I need/want to wear gloves, too?  My attempts to pull gloves on over the Flexmeters a couple of weeks ago were failures, as this would have required much larger gloves than I have available.  But, it turns out the other way around works fine, with the Flexmeters pulled on over the gloves:

20190516-205909524-i-OS.jpg

Nice!  With the jacket on, though, the Apple Watch becomes unusable tucked under the left wrist guard.  Hmm.

As for the ride itself, I felt very comfortable and in control.  Mounting is easy now that I've learned not to look down while doing it.  I even feel OK about standing in a turn lane or at an intersection, waiting for a car to pass, and launching during a gap before the next car arrives.  Did some speed bumps and almost flew off the wheel when I went over one too fast.  Gotta learn to do those little "lighten my weight" knee-flexed hops I've seen demoed in videos.  Might try on Sunday as the weather should be almost-dry that day.  Today for the first time, while stopping, I intentionally allowed the wheel to go into reverse for a few inches before dismounting.  First baby step toward eventually learning to go backward.

I'm ready for those summer work commutes, which means I wait for nicer weather or I decide to risk water infiltration.  The EUC Bodyglove would probably provide some protection, at least from direct rain impact.

Link to comment
Share on other sites

5 minutes ago, svenomous said:

Gotta learn to do those little "lighten my weight" knee-flexed hops I've seen demoed in videos.

That is an important skill indeed! I suggest practicing on actual curbs. Try to find a peaceful place where a curb gets progressively taller, a perfect place for practice.

Once you learn the actual curbs, scaling down to smaller bumps becomes effortless.

  • Like 1
Link to comment
Share on other sites

6 hours ago, svenomous said:

My attempts to pull gloves on over the Flexmeters a couple of weeks ago were failures, as this would have required much larger gloves than I have available.  But, it turns out the other way around works fine, with the Flexmeters pulled on over the gloves:

20190516-205909524-i-OS.jpg

 

You definitely want the wrist guards on the outside. The smooth plastic helps with the slide during a fall which you want to avoid jamming your arm and breaking something. 

  • Upvote 2
Link to comment
Share on other sites

9 hours ago, Dzlchef said:

You definitely want the wrist guards on the outside.

True, that is the ideal situation.  The other way around would be better than nothing, since the Flexmeter sliders are removable and under gloves they'd still function as wrist-guards (with limitation of flexion) and to a limited degree as impact absorbers.

Link to comment
Share on other sites

Today I finally commuted to work on the wheel, due to a rain break in the weather.  I was very apprehensive about riding in morning rush hour traffic, but the only way to get comfortable is to do it, so off I went.  I had a couple of embarrassing moments at intersections, where I didn't mount gracefully.  Mounting reliably under pressure is still my biggest issue, although I'm slowly getting better.  My second biggest issue is speed wobble.  Go a little too fast, or go over a bump just the right way, and wobble-wobble-wobble followed by a lean-back and some braking to dampen it.  I even had a braking wobble at one point that was very alarming as I was going downhill at pretty good speed (20mph) and suddenly and instantly it started a high-frequency wobble that was so fast it felt more like a strong rattle, and braking actually aggravated it, so I actually ended up accelerating a bit to dampen it.  Generally, standing very straight-legged reduces or eliminates the problem, but not always, and not entirely.  I've tried standing farther forward on the pedals (it seems to get worse), farther back (it doesn't go away entirely, and it's very tiring to stand too far back), farther outboard (when wobbles do happen, they seem worse), and with feet angled in or out.  I've tried bending knees to varying degrees and placing the knees farther forward or backward on the pads.  Nothing seems to quite work.  This is very frustrating.

Comfort (foot numbness/pain) also continues to be a problem, although it's slowly getting better.

Will probably do another commute on Thursday, as the weather looks like it will be acceptable.

As soon as I get a weekend opportunity (Memorial Day weekend seems like a candidate) I plan to do a long-distance ride on local trails, probably the "north Lake Washington loop" which is around 40 miles.  This is to increase endurance/comfort and continue to work on technique and mounting.

I think I'm at or near the 200km threshold that would allow me to increase speed limit, but there's no point in increasing it if I can't go faster than 15mph without wobbling.  Also, faster than 20mph is still not a good idea at my skill level.

  • Upvote 2
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...