Jump to content
Ken Taylor

Unicycle Batteries And The Design Flaw That Makes Wheels Dangerous

Recommended Posts

Shutdowns generally end like this and are as painful as this looks.

giphy.gif

Notice how her feet are in space and can't push against anything to break the fall. If, like me, you ride slower than you can run, overleans and other mishaps can be recovered by stepping off but the physics is different in a shutdown and you can't just step off. It is like this.

WheelForceDiagramCrashV2.png

This guy demonstrates the technique for an unpowered dismount. From the moment of unbalance he is falling but he lifts his knee quickly enough to get his foot forward under his body and uses the forward momentum to lift him like a pole vaulter. He knows it's coming and the physics is not opposite to what he is expecting so a shutdown for you is more difficult than for him. My four ended in more pain than any other mishap.

On many wheels battery protection has been prioritised over rider protection which can lead to a shutdown without warning. Wheels designed this way and fitted with low quality batteries can shut down at any time but even with good batteries they become dangerous as the batteries age. A rider is only likely to be aware of a rare and mysterious catastrophic failure on a wheel that was for a long time safe. Early battery replacement with good cells will avoid shutdowns or the battery can be modified.

The graph below for an Airwheel X8 shows published battery data overlaid with testing data of an aging cell X (black), damaged cells X (grey) and a generic battery with different, damaged cells X (blue). The discharge rate for the X tests was 0.2 amps. The aging cells were still able to sustain a voltage above the tilt back/warning voltage long enough to warn a rider but the damaged cells went from above the warning voltage to below the shutdown voltage too quickly. The generic battery had a shutdown voltage higher than the warning voltage, making it unsuitable for this Airwheel X8 from new.

8aafc2571fa58b89c5ffbab193ba8e5a7db2b040

A longer version of this summary is published elsewhere.

  • Like 1
  • Upvote 1

Share this post


Link to post
Share on other sites

@Ken Taylor

Welcome to the Forum...

Seams you where riding a bit longer on this airwheel? Now that you are here, you can have a lookout to “real good” electric-unicycles, with multiple parallel batterie systems (up to 8, 1680wh), better BMS’s and much more power and speed! Real nominal 1500Watt, safe riding on 30kmh or even faster....

Airwheel has missed the train a long, long time ago...i guess since over two years they havent put out a new model. 

Edited by KingSong69
  • Like 1

Share this post


Link to post
Share on other sites

First day riding I met a woman who was out on an airwheel X5 who showed me a scar across her forehead and told me the knocked out nine teeth when she faceplanted off of her wheel a year ago.  Apparently her wheel was broken and airwheel gave her a hard time about replacing it and she had to prove it had been a mechanical failure. 

  • Like 1

Share this post


Link to post
Share on other sites
4 minutes ago, mezzanine said:

First day riding I met a woman who was out on an airwheel X5 who showed me a scar across her forehead and told me the knocked out nine teeth when she faceplanted off of her wheel a year ago.  Apparently her wheel was broken and airwheel gave her a hard time about replacing it and she had to prove it had been a mechanical failure. 

;)

Share this post


Link to post
Share on other sites

Thanks for the responses. It is sad that

On 10/2/2017 at 4:29 PM, KingSong69 said:

the linked article did not help the e-wheel community very well ;-)

Until I understood the problem I didn't know the right search terms. I can now find lots of mentions but I've not found explanations of why it is worse than a normal mishap and three quarters of what I can discover is confusing or wrong. I can find no other comprehensive explanation and even now I can't find it explained how

On 10/2/2017 at 4:29 PM, KingSong69 said:

top quality brands like GW, KS, Inmotion, Ips are not designed like this anymore.

with the exception of seeing it suggested that KingSong wheels do not have switches in the BMS to disconnect the battery from the main board. How is it solved in the other brands, have wheel manufacturers abandoned low voltage battery protection entirely? Have they implemented the solution I suggested under the heading What Can Manufacturers Do? . If I go to buy a wheel BMS from Aliexpress I see they have low voltage shutdown and if I buy a replacement battery they have low voltage shutdown. One of my shutdowns was inevitable because the replacement battery had the low voltage shutdown set higher than the X8 tilt back voltage, which must happen to others as well. You have to have a good understanding of the issues to even think of checking this. 

On 10/2/2017 at 4:44 PM, KingSong69 said:

Airwheel has missed the train a long, long time ago...i guess since over two years they havent put out a new model. 

Yes, but as far as I can tell based on numbers of eBay listings, Airwheels and generics are still the best selling. They will all become dangerous as they age and generics with low quality batteries are dangerous from first use because of the design flaw. I started out clueless about wheeling and didn't want to spend too much until I'd tried it. Generics and Airwheels are priced for that type of buyer.

It was a surprise to me when the X8 shutdown and wasn't obvious, to me at least, why. I expect situations like

On 10/2/2017 at 4:51 PM, mezzanine said:

First day riding I met a woman who was out on an airwheel X5 who showed me a scar across her forehead and told me the knocked out nine teeth when she faceplanted off of her wheel a year ago.

to become more commonplace as the number of aged wheels grows. When you've been riding comfortably for a long time you don't expect a design flaw to have been there from the beginning that will inevitably hurt you if you ride long enough, so it shocked me. They are still being sold new and the design flaw is a bit like the Takata airbags problem on a smaller scale.

Also, being rather cowardly, I wouldn't consider

On 10/2/2017 at 4:44 PM, KingSong69 said:

safe riding on 30kmh or even faster....

to be possible because mishaps can happen and once the wheel speed is faster than running speed, falls would be difficult to avoid and at that speed hurt. Being able to step off is great for safety. Overleans are common on my X8 but OK when stepping off is easy.

Share this post


Link to post
Share on other sites

Screenshot_20171008-001358-01.thumb.jpeg.4a01d19890d22618643bcca70dd50a9c.jpeg@Ken Taylor

On 10/2/2017 at 7:29 AM, KingSong69 said:

This goes only for old wheels, generic china clones and non reputable brands....

Orbtronic 18650 PD are Panasonic and they are the same quality as GW, KS battery cells if not exactly the same cell.

The only mistake of Airwheel is to provide small capacity li-ion battery pack : main issues are lack of power and faster aging because more cycles are needed than bigger batteries.

As @Ken Taylor showed there is a voltage drop using 18650 li-ion technology (his graph could be used for some major brand also because it is voltage related ). Other solutions are :

- bigger battery packs (340Wh, 680Wh etc.), 170Wh should  be forbidden

- use of li-po (but they are rare) which are able to deliver more current. I had a 180Wh, strong acceleration and battery was often warm after a ride

- use of LiFePO4 because the voltage do not drop as much as li-ion so no need to compensate with greater current. Greater cycle life.

- last option (not best) is to use more recent 18650 battery cells from Panasonic for example. As a comparison, Panasonic 18650PD vs 18650GA (able to deliver 15A sustained also) used in Mten3.

  Screenshot_20171008-001358-01.thumb.jpeg.4a01d19890d22618643bcca70dd50a9c.jpeg

Edited by jbwheel
  • Upvote 1

Share this post


Link to post
Share on other sites

@jbwheel 18650GA is only rated 10Amp continous...and beside of that it is used in nearly every wheel from KS and GW...for example, Acm, Msuper, Ks16s, ks18s...

 

Share this post


Link to post
Share on other sites
3 hours ago, KingSong69 said:

@jbwheel 18650GA is only rated 10Amp continous...and beside of that it is used in nearly every wheel from KS and GW...for example, Acm, Msuper, Ks16s, ks18s...

 

What I think is a good approximation (at least for GW and some KS) : packs rating multiple of 340Wh are Panasonic NCR18650PF (very similar to PD).

There is no way to limit GA to 10A. It has been measured at 15A (test stopped because of the heat). It will be a problem for the new wheels displaying 512Wh (Mten3, MCM4 PRO...), compared to previous 680Wh. 2 packs instead of 4 means much faster aging, more than 2 times because of the higher currents and the generated heat (also means less autonomy)...

Screenshot_20171008-160814-01.thumb.jpeg.79da1779dff7d4eeadb5b85a341cac03.jpeg

Edited by jbwheel
  • Like 1

Share this post


Link to post
Share on other sites
1 hour ago, Slaughthammer said:

As for the 512Wh wheels, those batteries are perfectly capable of delivering 1,48 kW cont

They are able to cope but they will age much faster than my MCM4 680Wh for example.

1 hour ago, Slaughthammer said:

high temperature, that tells me that the cell is not suitable for cont. 15A draw. 

1 hour ago, Slaughthammer said:

 

That's the point, it is how they are used. Also, you can can see that the test was interupted after about 2/3 capacity was used at 15A which gives some margin that GW uses as a nominal. Then it depends if continuous means more than 10s or so (usual definition, contrary being peak) or a full run at 15A...

1 hour ago, Slaughthammer said:

400W delivered from the battery, which translates to a moderate 2,7A per cell.

Not so simple, maybe because it is 10" or the motor is too powerful or not efficient at low/medium speed. I have already started to do some measures, below is some results I have shared for downhill / flat. But I do every day the same uphill (about 1-2min 10% at 20-25km/h), it is quite demanding.

59da7098596c2_DescenteP.Versailles25sept.thumb.jpg.59ff57b8e2a991af4e42a6e4b17c75c4.jpg

  • Like 1

Share this post


Link to post
Share on other sites
15 minutes ago, KingSong69 said:

@Slaughthammer i leave that discussion to you :rolleyes:

How generous of you... :cry2:

@jbwheel yes, aging is a little faster with higher amps drawn. But still the main factor for battery aging remains cycle depth, especially when the amps drawn are quite well within the specified capabilities of the battery. If it takes you longer than an hour to empty the battery, the average consumption is below 1C... 1C in this case translates to 3,5A/cell. The cells are specified for up to 10A continuous (with continuous meaning constant draw until the battery is empty!). No battery manufacturer specifies a peak amperage, so that's left to guesswork. 

As for that diagram, how were those figures collected? Let me guess, you took a smartphone app like wheellog or darknessbot... those figures are on the one hand unreliable and on the other hand it's the motor current, not the battery current. As the motor controller works as a stap down converter, the amperage on the motor side is, depending on the speed, much higher than on the battery side. 

And with that beeing said, I'm outta here...

Edited by Slaughthammer
  • Like 1

Share this post


Link to post
Share on other sites
3 hours ago, Slaughthammer said:

If it takes you longer than an hour to empty the battery, the average consumption is below 1C... 1C in this case translates to 3,5A/cell. The cells are specified for up to 10A continuous (with continuous meaning constant draw until the battery is empty!)

I don't know if you are a seller but you are very misleading. Aging is not a matter of averaging but how much actual stress (xC) you put on the battery (typical for li-ion is 500cycles, 1C). GA max discharge rate is 10A (~3C) but it is able to deliver in a continuous steady state 15A (electrical definition, 4C+) which 18650PF is not able to do. Using it like that (strong acceleration, uphill, ...) IS still stressfull when a good li-po would not even feel it (not mentioning cold weather conditions).

As a consequence, peak power for GA (a fraction of second to 10s following A123 definition) is probably much higher, reducing faceplant probability.

For the diagram, you are right, it comes from the wheel. Which is still an indicator since @LEE4ERmeasured with a ampmeter 30A battery side,  on a less powerful wheel (MCM4) pressing it  just with his hands. See also other results obtained with eagle tree logger etc.

 

  • Like 1

Share this post


Link to post
Share on other sites
On 02/10/2017 at 7:07 AM, Ken Taylor said:

This guy demonstrates the technique for an unpowered dismount.

I would recommend to anyone, and in particular anyone in fear of wheel failures, to train unpowered mounting on a regular basis. It is somewhat less difficult than expected. I have been practicing cutting off the wheel under me at low speed, but it turns out to be very quickly a rather boring exercise. Maybe I should go back to my old IPS to be able to do this at higher speed as well. 

Share this post


Link to post
Share on other sites
On 10/10/2017 at 7:52 PM, Mono said:

I would recommend to anyone, and in particular anyone in fear of wheel failures, to train unpowered mounting on a regular basis. It is somewhat less difficult than expected. I have been practicing cutting off the wheel under me at low speed, but it turns out to be very quickly a rather boring exercise. Maybe I should go back to my old IPS to be able to do this at higher speed as well. 

How do you accomplish this exactly?  As far as I know it's not possible to turn off a wheel while it's moving.

Thanks.

  • Like 1
  • Upvote 1

Share this post


Link to post
Share on other sites
1 hour ago, Marty Backe said:

How do you accomplish this exactly?  As far as I know it's not possible to turn off a wheel while it's moving.

Thanks.

He would have to install some sort of switch,either cutting board power or a 3pole disconnecting the motor wires.

  • Upvote 1

Share this post


Link to post
Share on other sites
46 minutes ago, Jrkline "Wheel Whisperer" said:

He would have to install some sort of switch,either cutting board power or a 3pole disconnecting the motor wires.

I agree, yet I have my doubts that's what he's doing.

Share this post


Link to post
Share on other sites

Thanks for the reply   @jbwheel I wasn't concentrating on battery life  but safety issues though I did comment on distance per charge in the full article which a bigger battery would disproportionately improve.

I've added Airwheel X8 tilt back and cut off voltages in your battery graph. It looks safe to me as there is a good horizontal distance between tilt back and shut down. The problem comes when the graph goes vertical as I've illustrated with the added curve marked "dangerous"which happened with damaged batteries for me  but will happen with all batteries when they get old enough. Then shut down happens without there being time for a tilt back.

Screenshot_20171008-001358-01-2.png

  • Like 1

Share this post


Link to post
Share on other sites

 @Mono it never occurred to me to practice unpowered dismounts. As @Marty Backe said, I don't know even know how to do it. I've tried starting unpowered a few times but never been able to mount successfully. I thought it was impossible.

Edited by Ken Taylor
grammar
  • Like 1

Share this post


Link to post
Share on other sites
15 hours ago, Ken Taylor said:

 I wasn't concentrating on battery life  but safety issues though I did comment on distance per charge in the full article which a bigger battery would disproportionately improve.

Battery aging is worsening the situation, like cold temperature...

As said before, the good choice is a battery which high current capability so that voltage would not drop (lipo like @EUC Extremedid or good LiFePO4) when current need is multiplied by 3 or 4 = consistent behavior from the beginning to the end.

Some manufacturer did chose lipo/lifepo4 but were not commercially successfull as nobody tests wheel limits at 20% battery or by cold weather. Onewheel on the contrary seems to do well but because of its original design.  This is not new nor a battery revolution as some people may think. Also LiFePO4 aging is at least 3-4 times better making it much more ecological...

A123-discharge-curve.jpg

Edited by jbwheel

Share this post


Link to post
Share on other sites
39 minutes ago, jbwheel said:

This is not new nor a battery revolution as some people may think. Also LiFePO4 aging is at least 3-4 times better making it much more ecological...

Even more important, where LiFe cells are concerned, is the very high current capability with tiny drop in voltage, meaning you do not need massive packs to get safety out of LiFe cells they will hold their voltage until virtually empty. As well, of course having massively higher cycle life. So they are absolutely ideal for “last mile” light weight commuting rather than going miles and miles on the things.

Several wheel manufacturers have used LiFePO4 batteries, including Solowheel and Uniwheel. Universally, at least within this forum, users have complained vociferously about the cost versus poor capacity. So, bottom line, people do not want to pay the price that short range LiFe wheels cost.

IMHO, cold weather is every bit as dangerous as aged batteries and, actually, does an awful lot of aging itself. I keep my model aircraft batteries in a heated bag in winter now.

 

  • Like 1

Share this post


Link to post
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...