Future EUC Wishlist

[zeke]

Gonna have to add another one to this list.

As discussed here, it's a big problem that we don't know how much thrust margin we have until we faceplant. We're given speed beepers, but they are programmed to activate when we have some arbitrary and unknown amount of thrust margin left, which is truly only appropriate for a single, and unknown, riding condition. Many people have fallen well before hitting the beeper, because they made large a demand for thrust that exceeded what the EUC could provide at that speed and state of charge. My boss at work fell off his OneWheel for the same reason: on a low state of charge, he tried to accelerate and instead of tiltback, he got nose dive.

Where D is the motor drive duty cycle ranging from 0% to 100%, the thrust margin is proportional to (100%−D). Having an audible chirp that repeats with a period C(100%−D)/T (where C is a proportionality constant and T is the EUC's absolute temperature in Kelvin) would allow us to develop an intuitive feel for how close we are to the edge so that when we get near it with repetition and experience, we can better judge where it is and avoid it.

Edited November 18, 2019 by zeke

[Aneta]

35 minutes ago, zeke said:

Gonna have to add another one to this list.

As discussed here, it's a big problem that we don't know how much thrust margin we have until we faceplant. We're given speed beepers, but they are programmed to activate when we have some arbitrary and unknown amount of thrust margin left, which is truly only appropriate for a single, and unknown, riding condition. Many people have fallen well before hitting the beeper, because they made large a demand for thrust that exceeded what the EUC could provide at that speed and state of charge. My boss at work fell off his OneWheel for the same reason: on a low state of charge, he tried to accelerate and instead of tiltback, he got nose dive.

Where D is the motor drive duty cycle ranging from 0% to 100%, the thrust margin is proportional to (100%−D). Having an audible chirp that repeats with a frequency CT(100%−D) (where C is a proportionality constant and T is the EUC's absolute temperature in Kelvin) would allow us to develop an intuitive feel for how close we are to the edge so that when we get near it with repetition and experience, we can better judge where it is and avoid it.

Chirping feedback will be VERY intuitive, akin to chirping barometric altimeters that glider pilots use to find the thermals (the frequency and tone indicate rate of ascent).

Curious: why proportional to temperature? Decreasing battery capacity? Or to make the more sad chirping as the temperatures go lower for those who live in real winters and watch @Marty Backe's videos from California with envy?

[zeke]

1 minute ago, Aneta said:

Chirping feedback will be VERY intuitive, akin to chirping barometric altimeters that glider pilots use to find the thermals (the frequency and tone indicate rate of ascent).

Curious: why proportional to temperature? Decreasing battery capacity? Or to make the more sad chirping as the temperatures go lower for those who live in real winters and watch @Marty Backe's videos from California with envy?

Haha! That was just a quick stab at compensating for the increasing winding and MOSFET resistance with temperature.

[zeke]

10 minutes ago, Aneta said:

Chirping feedback will be VERY intuitive, akin to chirping barometric altimeters that glider pilots use to find the thermals (the frequency and tone indicate rate of ascent).

Curious: why proportional to temperature? Decreasing battery capacity? Or to make the more sad chirping as the temperatures go lower for those who live in real winters and watch @Marty Backe's videos from California with envy?

Ideally the chirp period would scale with (1 − D)V_BATT/(R_ESR + R_MOSFETS + R_WINDINGS), so in cold climates where the battery ESR R_ESR rises rapidly, assuming the sum of the resistances to scale with absolute temperature is probably not my smartest idea.  Nonetheless, a simple implementation with chirp period equal to C(1 − D) would be better than what we have now. A more advanced implementation that estimates V_BATT/(R_ESR + R_MOSFETS + R_WINDINGS) would be best. 

[Aneta]

3 minutes ago, zeke said:

Ideally the chirp period would scale with (1 − D)V_BATT/(R_ESR + R_MOSFETS + R_WINDINGS), so in cold climates where the battery ESR R_ESR rises rapidly, assuming the sum of the resistances to scale with absolute temperature is probably not my smartest idea.  Nonetheless, a simple implementation with chirp period equal to C(1 − D) would be better than what we have now. A more advanced implementation that estimates V_BATT/(R_ESR + R_MOSFETS + R_WINDINGS) would be best. 

I think taking into account that human ear is not very precise Fourier spectrum analyzer, the feedback provided by simple formula would be sufficient. It's like weather temperature maps where they have a rainbow of colors representing temperatures from -40F to +100F, and then you look at any particular point and try to determine temperature by color. Perhaps, professional musicians can tell exact frequency of chirps, so for them there should be an option for temperature-dependent feedback.

[zeke]

12 minutes ago, Aneta said:

I think taking into account that human ear is not very precise Fourier spectrum analyzer, the feedback provided by simple formula would be sufficient. It's like weather temperature maps where they have a rainbow of colors representing temperatures from -40F to +100F, and then you look at any particular point and try to determine temperature by color. Perhaps, professional musicians can tell exact frequency of chirps, so for them there should be an option for temperature-dependent feedback.

EUCs already measure their temperature and battery voltage, so it shouldn't be difficult to throw in a lookup table to estimate V_BATT, R_ESR, and R_MOSFETS + R_WINDINGS. 

If the manufacturers upgraded their firmware with this, all EUCs everywhere would suddenly become a lot safer.

[xorbe]

13 hours ago, Jason McNeil said:

I'd like to a see a new version of the MTen3 with 4x modular battery packs, so in theory it could be taken aboard a plane.

This is money idea right here.

[zeke]

24 minutes ago, xorbe said:

This is money idea right here.

Yep I'd buy it in a heartbeat.

[Aneta]

The only reasonable way to have a battery with less than 100Wh capacity allowed by FAA in unlimited quantity without approval, made of cells of not ridiculously low capacity (like 1200mAh), is for it to be a no more than 10s1p configuration (36V). Similar to this outrageously expensive Grin battery:

https://www.ebikes.ca/shop/electric-bicycle-parts/batteries/36v-ligo-battery.html#

These can be assembled in pairs in series (to achieve "84V", in EUC terms) and then any number of parallels, to achieve any capacity, even Monster capacity! I guess one could custom design/3D print a shell that would take these blocks, as well as the motherboard from the standard EUC, then you just remove the controller and motor from your EUC, transplant into the new shell, take a suitcase full of these 100Wh blocks, and go fly!

 

[Aneta]

This got me thinking: what if our battery packs were not welded together, but were like how we load AA/AAA into flashlights and other gadgets, i.e. spring-loaded contacts. There are many projects on Endless Sphere where people build ebike batteries from 18650's like this. Yes, there are issues, like losing contact under heavy vibration, but they can be mitigated. But imagine the possibilities: air travel with any size EUC, anywhere in the world... Because each 18650 is only 10-13Wh, and number of them is not limited, one can take any capacity EUC anywhere on the planet.

Edited November 18, 2019 by Aneta

[mike_bike_kite]

@Aneta"take a suitcase full of these 100Wh blocks, and go fly!"

Wasn't there a good reason they won't allow large Lion batteries on air flights? Does splitting the battery into 4 separate units mitigate that risk?

I can imagine the headlines "747 brought down by monster EUC hidden in suitcase"

Edited November 18, 2019 by mike_bike_kite

[Aneta]

9 minutes ago, mike_bike_kite said:

@Aneta"take a suitcase full of these 100Wh blocks, and go fly!"

Wasn't there a good reason they won't allow large Lion batteries on air flights? Does splitting the battery into 4 separate units mitigate that risk?

I can imagine the headlines "747 brought down by monster EUC hidden in suitcase"

https://www.faa.gov/hazmat/packsafe/more_info/?hazmat=7

[Shad0z]

You forgot DURABILITY. Too many times you take a spill and go. Oh yeah, looks like I'll have to fix this. Or replace the shell. There should be removable bumpers in the design

[Mono]

25 minutes ago, mike_bike_kite said:

Does splitting the battery into 4 separate units mitigate that risk?

To some extend it does, it seems to me, as it takes (much) more time to propagate a fire from one pack to the next. But of course to a large extend it is also getting around the regulations and I can imagine they won't let you pass with 20 100Wh packs in your hand luggage. On a different perspective, given that every other passenger has a laptop with a 100Wh+ battery in the airplane, the plane with 200 passengers carries already more than 10 kWh of consumer batteries. Do we assume they imply the same risk as one single 10 kWh battery? The former risk seems to be extremely low, as we don't have a single incident of a plane we know of going down for that reason (AFAIK).

Edited November 18, 2019 by Mono

[Nic]

9 hours ago, Planemo said:

I just don't see the point of a built-in slow charger when the wheel will always be at home where a normal external charger would be.

Convenience, and the wheel might need a top up if visiting someone that is just a little bit too far for a round trip without top up. Sometimes you can change plans when out and need a charger, which you didn't bring because you weren't expecting to need one. Fast chargers reduce battery life and I wouldn't want to leave one unattended as fire risk is higher. A normal charger is perfect for overnight charging if you are sleeping over somewhere or on vacation. Many of us don't like having to carry a separate brick around and we don't all have the same needs/demands you know.

[atdlzpae]

14 minutes ago, Nic said:

Fast chargers reduce battery life and I wouldn't want to leave one unattended as fire risk is higher.

The effect of reduced battery life is negligible. Even a 10A charger is nowhere close to destroying the battery.
Fire risk depends on the build quality, not on the amperage of the charger.

@Aneta This could definitely work! Since there is no limit up to 100Wh, even MSX could be made modular with only 16 modules. 

Edited November 18, 2019 by atdlzpae

[Nic]

12 minutes ago, atdlzpae said:

The effect of reduced battery life is negligible. Even a 10A charger is nowhere close to destroying the battery.
Fire risk depends on the build quality, not on the amperage of the charger.

Who said anything about destroying the battery? And is this just your 'opinion' or do you have some figures to back up your assertions? Current is what causes heat and that is greater when fast charging is it not? Are you certain 100% that your battery is perfect? Does it remain in perfect condition throughout its life? If everything is so great with fast charging they why isn't this the new normal? Some facts please. 

Edited November 18, 2019 by Nic

[atdlzpae]

@Nic For example, Gotway 84V 1600Wh is using configuration 20s6p.

6p at 10A = 1.6A per 18650.
According to AliExpress they are using 3.5Ah NCR18650GA.
According to the datasheet if you charge NCR18650GA with 1.6A from 0% to 100% (nobody rides from 0 to 100%... damn tiltback), you'll still have (page 4 of the datasheet):
- 82% capacity after 100 cycles (6000km)
- 74% capacity after 200 cycles (11000km)
...
- 62% capacity after 500 cycles (25000km)

And there are no fast chargers for MSX faster than 5A. So the fastest fast charger (within spec) will have negligible impact on your wheel overall.

 

1 hour ago, Nic said:

Current is what causes heat and that is greater when fast charging is it not? Are you certain 100% that your battery is perfect?

By "destroying" I meant "destroying the capacity". A damaged battery is a fire hazard regardless of how slow your charger is.
Fast charging is not normal because:
1) Most people don't need fast charging
2) Fast chargers cost more

Edited November 18, 2019 by atdlzpae

[JZT-Colorado]

4 hours ago, xorbe said:

This is money idea right here.

Damn right!!

[Nic]

2 hours ago, atdlzpae said:

@Nic For example, Gotway 84V 1600Wh is using configuration 20s6p.

6p at 10A = 1.6A per 18650.
According to AliExpress they are using 3.5Ah NCR18650GA.
According to the datasheet if you charge NCR18650GA with 1.6A from 0% to 100% (nobody rides from 0 to 100%... damn tiltback), you'll still have (page 4 of the datasheet):
- 82% capacity after 100 cycles (6000km)
- 74% capacity after 200 cycles (11000km)
...
- 62% capacity after 500 cycles (25000km)

And there are no fast chargers for MSX faster than 5A. So the fastest fast charger (within spec) will have negligible impact on your wheel overall.

 

By "destroying" I meant "destroying the capacity". A damaged battery is a fire hazard regardless of how slow your charger is.
Fast charging is not normal because:
1) Most people don't need fast charging
2) Fast chargers cost more

Thanks for clearing that up.

You do realise that the datasheet says nothing about cycle life for different charging rates? ... its just data for discharge rates.

[Planemo]

2 hours ago, Nic said:

Convenience, and the wheel might need a top up if visiting someone that is just a little bit too far for a round trip without top up. Sometimes you can change plans when out and need a charger, which you didn't bring because you weren't expecting to need one. Fast chargers reduce battery life and I wouldn't want to leave one unattended as fire risk is higher. A normal charger is perfect for overnight charging if you are sleeping over somewhere or on vacation. Many of us don't like having to carry a separate brick around and we don't all have the same needs/demands you know.

Re: 'top up' lets take for example an 1860Wh MSX, which is a fairly average size battery these days. The slow charger (1.5A) will provide around 6% battery per hour. That had better be a very large cup of coffee you have at your visit for your top up to make any difference. This point also applies to your 'change plans when out and need a charger'.

Re: fast chargers reducing battery life/increasing fire risk, no they don't as has been explained well in posts above.

Re: overnight stays yes slow chargers are great but given that slow chargers are small and light, for me it's not an issue to just pack it with everything else . I guess it also depends how many times you take your wheel with you for overnight stays and/or vacations.

Re: not having the same needs and demands, I get that, which is exactly why I added 'YMMV' to my last post. I was just trying to give a different perspective on how useful (or not) an onboard slow charger is. As you say, we all have different needs.

I will make it clear that I am not against onboard chargers. The point I am trying to make is that EUC batteries are getting exponentially bigger, and so for an onboard charger to be worthwhile (to me) it needs to be high amperage. This means big components. If the manufacturers can make it work in the confines of an EUC then great.

I can see an argument (possibly) for onboard 1.5A chargers on small (say 320Wh) wheels but if you are really using the onboard facility often then you would have to question why you wouldn't just buy a bigger wheel and avoid the inconvenience of range anxiety/charging in the first place. Maybe the last miler crowd could benefit where say a V5F is great, portable, and could get a (relatively) worthwhile battery gain in an hour at the coffee shop but given how many people are buying 1000Wh+ wheels these days I would say it's quite a small market.

[Mono]

About half of my chargers don't work as they are supposed to. As it is, I don't care, but if it were the onboard charger I would. An onboard charger would not only need to be much more reliable than any typical charger is, but it would also have to withstand much much more mechanical distress with the high reliability. I am not sure the benefits of an onboard charger would be worth the additional costs.

[Nic]

5 minutes ago, Planemo said:

Re: fast chargers reducing battery life/increasing fire risk, no they don't as has been explained well in posts above.

https://www.theregister.co.uk/2019/06/05/liion_battery_scan_charging/

Quote

 

Boffins at Purdue University in America say they made the discovery while conducting atomic-level scanning of fast-charging lithium-ion battery electrodes. The team, led by Kejie Zhao, an assistant professor of mechanical engineering, used multiple scanning techniques and computer-aided enhancements to take a deep dive into a particular power pack, and found what may be the Achilles' heel of fast charging.

Essentially, what happens is that fast charging seemingly damages the battery's electrode particles, forcing the device to lose capacity and shorten its lifespan. This degradation is caused by lithium ions whizzing between internal electrodes, we're told.

 

 

7 minutes ago, Planemo said:

I will make it clear that I am not against onboard chargers. The point I am trying to make is that EUC batteries are getting exponentially bigger, and so for an onboard charger to be worthwhile (to me) it needs to be high amperage.

Sure, the batteries are getting bigger, but the miles per watt-hour is still roughly the same so an hour on a slow charger might be enough to make sure you can make it home. This is better than not having a charger don't you think?

You already explained that you don't need a built-in slow charger, that is fine. Others may have a different view. We don't all have to agree on what we would like to see on our wishlist.

[Planemo]

14 minutes ago, Nic said:

You do realise that the datasheet says nothing about cycle life for different charging rates? ... its just data for discharge rates.

Standard charge rate for Sanyo NCR18650GA is 1,67A. So even at 10A charge, the MSX used in the example is within manufacturers charge recommendations.

Point being, 10A isn't even 'fast charge', it's 'standard' charge.

[Nic]

1 minute ago, Mono said:

About half of my chargers don't work as they are supposed to. As it is, I don't care, but if it were the onboard charger I would. An onboard charger would not only need to be much more reliable than any typical charger is, but it would also have to withstand much much more mechanical distress with the high reliability. I am not sure the benefits of an onboard charger would be worth the additional costs.

What additional costs? A simple board that is easily replaced would probably cost less. But if you are saying that chargers aren't reliable, then that is a worry, but even with an external charger you would still need to source a replacement. And why not both?... built-in slow charger plus external port for fast charge? I haven't had any charger fail yet for my laptops and these things only run for a few hours a day at most when charging, so if quality components are used I would expect the charger to outlast the battery. @Jason McNeil would probably have a better idea on charger reliability as he will have to deal will replacements ... not talking about upgrades here.