Live from EuroBike 2018

[R120]

EMBN Vid - Box 9 Speed EMTB drivetrain looks interesting, with beefed up chain

 

[R120]

Really like the look of this new Centurion 29er:

Centurion Numinis Trail E – 29er eMTB with broad field of application | E-MOUNTAINBIKE Magazine

 

[Rob Rides EMTB]

This was interesting.

Brose Drive S Mag motor, still in a prototype phase by the company Maxx.

 

[Rob Rides EMTB]

How about some ABS on your bike?!

 

[eFat]

Brose Drive S Mag motor, still in a prototype phase by the company Maxx.

Are you sure it's the "Mag"? The frame and plastic casing are exactly the same as previous version.

 

[Rob Rides EMTB]

Are you sure it's the "Mag"? The frame and plastic casing are exactly the same as previous version.

 

[Mabman]

Did you find anything on the long lasting chain front?
Any Range extenders from BMZ?
Any mention of LifePO4 cells being used?

I am not at Eurobike but have been following the progression of battery technology for a number of years. My first eBike in 2001 had a SLA one which was a really poor performer but peaked my interest about eBikes enough so that I have kept my eye on them. Batteries are the heart of any eBike and the stronger the heart the better. Although the focus is more on watts and nm of the motors in order for them to achieve these numbers it is just as much about the ability of the battery to supply the necessary energy.

Pretty much across the board, at least from the major manufacturers LiFePO4, commonly referred to as IFR cells, were replaced several years ago with LiMn204 IMR cells.

As far as range extension BMZ is supplying these cells to manufacturers: https://www.bmz-group.com/dokumente/BMZ-21700-50EL_36629.pdf They offer benefits over the 18650 cells they will eventually replace across the board and have extended range capability due to their across the board ratings as seen below in a simple comparison I put together awhile ago.

The maximum continuous discharge rate, or C rate, is the key here. It is over twice as much and that allows one to have the same performance using a lower rate of discharge which will extend range. Plenty of info about C rate about btw if that concept is not understood. It is one of the basic principles of battery tech that has the most to do with their overall performance and why BMZ, Tesla, Sony, LG, Sanyo etc. are going to the larger cell.

Some may notice that the comparison is between a lower end of the 18650 mah spectrum but even the strongest 3600mah ones don't compare to one with 5000mah and you can extrapolate from there.

No one at this time that I know of that is making packs with these cells to match up with all the various types out there, of which when you think of it are many due to each manufactures having to do things their own way and the proprietary nature of the firmware they have. Some cross over between brands but not many outside of Open Source ones like the Hai Long type typically used in the kit market place. Some enterprising hacker could figure out a way around the firmware issue but to get a battery to market each one has to undergo about a $15k UN standard testing procedure and development costs can be just as high.

However pretty much all the higher end models presented at the show have the new cells on board and especially in the Haibikes case is what allows them to achieve such a high nm rating and stated 10A charge rate. Although if you look at the info above the BMZ cell on the rights maximum charge rate is 7A so there must be an even stronger battery out there to handle 3 more A of charge that I am not aware of.

The question is why, as they have been in development for the last 3 years and in mass production for the last 1 year, and create the performance figures that the manufacturers use as marketing tools aren't they even mentioning them in their sales pitches?

 

[eFat]

It's well hidden, maybe I could retrofit mine!

 

[Doomanic]

The question is why, as they have been in development for the last 3 years and in mass production for the last 1 year, and create the performance figures that the manufacturers use as marketing tools aren't they even mentioning them in their sales pitches?

Existing supply contracts?
Longevity trials?
Or just to annoy eBikers?

 

[R120]

Its the frame development in bicycle that takes time, the technology may exist but it take a few years for manufacturers to go thought the design and prototyping of the frames to accommodate it.

 

[Doomanic]

That doesn’t explain Bosch and Shimano with their external packs though.

 

[R120]

Elf and safety? I presume getting a new battery on the market requires a lot of in field testing and lab testing

 

[Mabman]

Existing supply contracts?
Longevity trials?
Or just to annoy eBikers?

Considering the fact that electric cars take between 2,500-5,000 cells per car manufacturing and supply line is in place, contracts for them must be in the millions for some accounts. However a 48v 15ah 21700 battery only takes 39. So even if you are making a lot of batteries it doesn't add up to much. They are available in small quantities though via getting them from vape suppliers.

Initial reports I read a few years ago forecasted upwards of 2,000 charge cycles. Up to 7A charger and a long lunch to 85% and it's another 40 miles of smiles.

Hard be annoyed if you don't know what to be annoyed about and less annoying perhaps than looking forward to having a better battery in the near future as a replacement option when yours starts to degrade, or the enhanced performance will get your wallet out as soon as they appear and you'll now have two batteries and can skip the long lunch by stashing it at the lunch spot.

 

[Rob Rides EMTB]

Yes the Alpine E - i know its been out this year but so little info on it - might be a good one for you to do a test ride video on when you are back in UK?. The hardtail really interests me though, i saw some info on it on Singletrack.

View attachment 2941

View attachment 2942

View attachment 2943

I spoke with Orange about the HT. they aren’t sure if it’ll ever go to market. They feel the price point is too high compared to many others (and the can’t get it any cheaper).

 

[R120]

Its difficult because i cant see a better HT E-Mtb out there than the Vitus E-Sentier VR at 2.8k - unless you want Carbon nothing comes close in price and spec, and its based on a multiple award wining hardtail. Guessing orange would be at least 4k for same spec. Though as usual with Vitus no bloody water bottle bosses!

Vitus E-Sentier VR Hardtail E-Bike - SLX 1x11 2018 | Chain Reaction Cycles

 

[Rob Rides EMTB]

I think they were suggesting £3800 for that Orange HT IIRC

 

[R120]

Thats not too bad actually

 

[Kiwi in Wales]

Considering the fact that electric cars take between 2,500-5,000 cells per car manufacturing and supply line is in place, contracts for them must be in the millions for some accounts. However a 48v 15ah 21700 battery only takes 39. So even if you are making a lot of batteries it doesn't add up to much. They are available in small quantities though via getting them from vape suppliers.

Initial reports I read a few years ago forecasted upwards of 2,000 charge cycles. Up to 7A charger and a long lunch to 85% and it's another 40 miles of smiles.

Hard be annoyed if you don't know what to be annoyed about and less annoying perhaps than looking forward to having a better battery in the near future as a replacement option when yours starts to degrade, or the enhanced performance will get your wallet out as soon as they appear and you'll now have two batteries and can skip the long lunch by stashing it at the lunch spot.

Thanks for the heads up on this. It is very informative and it is what I have been looking into for a few months and thinking that I really do not understand why more manufacturers are not using these cells more. It’s a no brainer for me.

The 21700 (21mm x 70mm) cells are only slightly bigger than the 18650 (18mm x 65mm)
5000MAh v 3400 to 3500MAh
Possibly up to 1500 to 2000 v 800 charge cycles. 700 as stated by Specialized on their Mission Control App

If you could hack your way around the brains present in the current Levo/Kenevo battery you could possibly repack with 21700 batts as long as the extra 3mm height and 5mm width of the cells fitted giving you a 720Wh battery. If space was an issue which it could well be.... you could re orientate the cells by 90 degrees and 3D print a new enclosure which would create a small hump out the front of the down tube but may also allow you to slot in a few additional cells giving you an even larger value Wh battery?

Would be good to get your thoughts on the above?

As @ccrdave said in another thread, things can be adapted with some smart engineering.

 

[Doomanic]

You have to why why, it Vitus can get it under £3K, why can't Orange?

 

[R120]

Because the CRC empire can get the parts for cheaper - Vitus already mentioned that the reason their E-Bikes are such good value is they struck a great deal with Shimano for the E8000 system

 

[Doomanic]

Yeah, I get the whole economies of scale, but £1000 cheaper? Is the Orange better specced at that price?

Vitus is tiny compared to the big boys, I do think that some of the manufacturers are trading on their name and overcharging for bikes in general.

 

[Rob Rides EMTB]

Beauty or beast?

 

[Mabman]

Hey Kiwi,

As far as I can figure never having had one opened up the Levo standard battery is a 12s/3p. 12 cells long by 3 of those high.

12s x 3600mAh (usually arrributed to 18650 cells of high caliber) = 4.32ah

12s x 3p = 12.96ah (36 cells total)

12.96ah x 36v = 466.56wh Close enough for government work....

In order to get a comparable ah pack with the 21700 cells, that although they say 5000mAh nominal capacity are 48500mAh rated capacity so, you would need a 9s3p pack.

9 x 4850mAh = 4.365ah

9s x 3p = 13.095ah (27 cells total)

13.095 x 36v = 471.42wh

The new pack would be 189mm x 71mm x 63mm. I don't have a Levo battery to compare that figure to however but based on conventional 18650 sizing would be around 216mm x 65mm x 54mm. New battery would be shorter but wider and taller. You might be able to stuff them in your housing but it would take some creative placement of the cells perhaps but I doubt it. The extra width is a deal killer I've found for putting them into standard kit battery casings at least.

In order to get to 720wh +/- you would have to group the cells in 10s4p which would give you a 19.4ah battery x 36v = 698.4. with a total of 40 cells minimally sized 210mm x 71mm x 84mm. The 10s4p battery should be better in overall performance that a 720wh 18650 battery as explained next.

The big difference is that the new cell type battery would have a max continuous discharge rate of 42A. The best 18650 batteries max out at 30A. This means that the C rate is higher and so it puts less stress on the cells when you are only running the 15A or so that a Levo puts out gauging off volts x amps (current) is estimated to be 36v x 15A = 540w. This is what will transfer into better range in the end as the battery will draw down slower and provide the same amount of power throughout the discharge cycle. It can't produce more power unless you increase the controller A however in turn increasing watts.

As far as hacking into a proprietary controller here is a video explaining at least how it might work with a Bosch battery and would be surprised if it wasn't the same pretty much with all makers: Explanation starts at 26.32 in case it isn't qued up.

As noted by Justin, who is a wealth of knowledge and currently participating in SunTrip that started in Lyon and ending up in Canton, grin.ca, all open source batteries just have pos/neg 2 pin connectors exiting them but those that want to keep their stuff less accessible are putting a third pin in the line, which is hooked up to the BMS that is .05v according to the vid and probably serves provides a signal for their proprietary communications download in the controller and BMS . If it is as simple as trying to figure out a way to keep feeding .05v to that wire then it would be a simple hack. There are voltage converters available in all sizes that could accomplish this by tapping into the pos/neg wires and then feeding into the third pin wire. Probably not that easy but it might just be.

Also end weight should be considered. The 21700 battery cells, 68.5g, weigh 3kg alone and add some grams for the building block cell holders, you can save here by hot gluing the cells together, wire/strip connectors, BMS and wires perhaps another 400 so 3.4kg. A complete 36v 20ah battery weighs 4kg +/-. Making a .6kg difference which incidentally is more weight than the new Brose Mag motor lost over last years model.

I've seen no hard numbers yet on estimated charge cycles but as said they should be more. Battery life with any battery is about good management meaning charging more than 85% and saving 20% at the other end. Also how many A you choose to charge at as more will slightly decrease battery life. 21700's promise more and that is good enough for me as I mainly am interested in the performance and extra range capability in a lighter package part.

And speaking of batteries: Bike EU

 

[Doomanic]

12S isn't 36V...

12 cells in series is 43.2V
9 cells in series is 32.4V

AFAIK, all Levo packs are 10s4p, only the individual cells capacity changes.

 

[Kiwi in Wales]

Thanks again for all the additional info.

Bit scary regarding the ebike battery warehouse fire

Please see the screen shot below. It is a right hand side on view of a Levo/Kenevo battery. 40 cells in total but I have no idea how it is configured. If I put a meter across the battery it is 41.4 volts. The battery weighs 2.96kgs and is long and skinny. 55mm x 75mm x 52cm long.

When you connect to the battery via the Specialized Mission Control when the battery is ‘removed’ from the bike but is turned ‘on’ it gives you quite a bit of info. I am not sure if the Bosch batteries do the same? Please see the other screen shot.

 

[Doomanic]

The cells used are each 3.6v nominal, when fully charged they’ll reach 4.1-4.2v.

10 cells in series gives you 36v nominal. 4 sets of 10 cells (in series) connected in parallel gives you the capacity.
Voltage times capacity gives you Wh.

 

[Mabman]

That is the same amount of batteries as the 10s4p example above in a different configuration. Doing the math 41.4v x 15A = 621w peak which lines up with what I have seen as an accurate # and is a 14.4ah/518.4wh. No way looking at the pic that you could stuff 21700 cells in there, the extra width is what will mess things up.

 

[Doomanic]

Are you trying to say the Levo battery is 20s?

 

[Doomanic]

That is a 20s2p battery 18650 (3.6v/cell) which is the same amount of batteries as the 10s4p example above in a different configuration.

Same number of cells, very different battery.
The 10s4p is 36v nominal, the 20s2p is 72v nominal! I don’t know of any eBike using a 72v battery pack.

 

[Mabman]

You are right Doom. It is a 10s, it's Saturday and weed is legal here is my excuse.