Custom 16" EUC with high-drain batteries and proper BMS by MonoCustomize/EUCyou

[yoos]

Monocustomize just built a custom 16" EUC and I am very excited as it's basically an ideal short-range city commute wheel: light and compact, but explosively powerful and fast thanks to high-drain Samsung 40T cells. I am not sure he will ever produce such wheels on demand, but to me this EUC is a proof of principle: it is absolutely possible to make a fast and torquey wheel in the compact and light-weight category (he gave a ballpark price for parts as over 1370$ and estimated for ordering such a wheel twice the price). (The wheel was made by DelightDevice, which is the same guy as MonoCustomize and EUCyou.). Apparently he hasn't tested top speed, but I would not be surprised if it would be the first sub 20kg wheel to hit 70kph. It already has hit 65kph.

This is also timely considering the Begode fire problems and how their 4p LG 50MT packs are inadequate for high-power EUCs.

Here are two videos about the build (part one and part two with basic test-drive at the end). You can find footage of later rides on the same youtube channel. 

Key specs

• Begode Nik+ 100V controller
• Tesla motor and SKF bearings
• 16"x 2.5" Kenda tire
• modified ACM shell with carbon armor in front and rear
• 24s2p packs of Samsung 40T with smart BMS with bluetooth module on each pack
• 88kph free spin

Some more details:

• lifted EUCyou hangers and other custom and reinforced parts
• EUCyou pedals with 4 spikes on chrome moly rods
• Nik headlight
• V5F trolley (extra 385g) can be attached

Edited September 3, 2022 by yoos
max speed update

[supercurio]

Thanks a lot for the full description!
I missed how interesting the video seris is due to the language barrier despite being a "Delight Device" subscriber. Didn't know he's also MonoCustomize and EUCyou either.

So high drain cells, yes finally! with smart BMS also..
681.6 Wh I guess, counting 24*2*14.2.

[yoos]

4 hours ago, supercurio said:

681.6 Wh I guess, counting 24*2*14.2.

Something like this. I am actually a bit confused about battery capacities on EUCs.

Edit: found some threads, indeed the cutoff voltage in EUCs is higher than 2.5V in EUCs so the 681.6Wh cannot be fully used. (That's true for all EUCs). Some 5-15% of the energy is reserved.

Edited November 15, 2021 by yoos

[div]

EUCs “0%” voltage is generally around 3.15V, with Gotway at 3.3V and the KS18XL at 3.0V.

Is not 2.5V the point after which the cell is considered damaged? Is that the min voltage used to calculate Ah?

Edited November 15, 2021 by div

[yoos]

36 minutes ago, div said:

Is not 2.5V the point after which the cell is considered damaged? Is that the min voltage used to calculate Ah?

More or less - the cut-off voltage is the voltage where the controller disconnects it from load to avoid possible damage. It does not mean that a battery is inevitably damaged if it reaches, say 2.1V, it's just that the chances of damage are too high at those voltages. It seems that the specified cut-off varies between chemistries, manufacturers, cells. Just like with other battery limits like max charge or max discharge it's a quantitative thing -- if you exceed the limits chances of damage/accelerated aging increase to unacceptable levels [which are ultimately also a matter of opinion], but the damage is not guaranteed. 

In case of EUCs it is not a trivial decision: should the controller let you overpower the wheel and have a high-speed faceplant or should it abuse the batteries for a short time to try avoid a crash. Apparently the consensus is shifting towards the former: the controller should let you faceplant, because the alternative is to create a potential fire hazard. I agree with this. Overpowers are usually the riders fault and an accepted risk and mostly endanger the rider alone. They are also predictable in a sense. Battery fires, on the other hand, can harm or kill multiple people, cause much more property damage and they are unpredictable -- you just now that there is a small chance of fire at any time (with increased chances when charging/discharging) and that's it. Fires also create much worse EUC publicity than faceplants.

Using high-discharge batteries looks like the obvious solution to move the weak link of the EUC from batteries to controller/motor. The battery is the only component that is a true fire hazard that can cause catastrophe when the wheel is not engaged/off. All other components -- fuses, boards, motors might fail, but they would do so during operation so the chance of an unattended fire happening is miniscule.

While high-discharge batteries generally have lower capacity, they also have a lower internal resistance/impedance which reduces heating and voltage sag. Still the ability to safely deliver high discharge alone is completely worth the slight loss of range.

[Planemo]

1 hour ago, div said:

Is not 2.5V the point after which the cell is considered damaged? Is that the min voltage used to calculate Ah?

Yeah most 'rated capacity' testing is done down to a 2.5v cutoff. So yes we do get a bit short changed on our EUC's when we have 3.15~3.3v as minimum. This 'available capacity' reduces further when you consider that most EUC manufacturers over rate the wheels to start with (Sherman is 3106Wh not 3200Wh for example) plus they also over rate the cells themselves given that the 3106Wh calculation is based on a 3500Mah cell but in reality the Sanyo GA cell is generally marketed as 3350Wh. Similar situation with other cells/EUC manufacturers.

Adding to that, maximum cell capacity specs are generally obtained at very low discharge rates. Again an example, the Sanyo GA gave 3345mah (slightly under spec as cell was only taken down to 2.8v) but this was only at a 0.2A discharge. Go up to 5A discharge and max capacity comes down 3208mah.

This is where high P packs earn their weight. Get the per cell discharge rate down by adding more cells in parallel and mileage will go up. This is why the Sherman does really well on range if it was compared to say a 6P setup. Not to mention the other benefits of high P configurations. Downsides are weight and cost of course.

In the end though, the main thing is that ALL the EUC manufacturers work pretty much the same way (over rating the capacities) so for us buyers, comparisons are generally easy when we choose a wheel. There are odd occasions, like when Kingsong rated their wheel correctly at 1554Wh (KS16X) and yet GW were rating their MSX at 1600Wh. GW aren't the only culprit though.

[supercurio]

4 hours ago, yoos said:

Something like this. I am actually a bit confused about battery capacities on EUCs.

Edit: found some threads, indeed the cutoff voltage in EUCs is higher than 2.5V in EUCs so the 681.6Wh cannot be fully used. (That's true for all EUCs). Some 5-15% of the energy is reserved.

Yes I quoted a spec (down to 2.5V) for a fair comparison, since other manufacturers also show the cell capacity for the pack - vs what the wheel can actually drain.
This test: found when down to 3.2V:  11.3Wh (10A continuous) and 10.1Wh (20A continuous)

If it's Delight Device riding the wheel in 2P configuration, the usable capacity will in between to 24*2*11.3  or 10.1   (542.4 and 484.8 Wh respectively)

Certainly not a long range wheel but might be super fun as last mile device in the city, for tricks and play in skate parks also.

[DavidB]

Not sure what would be the advantage over his modified Teslas.  Looks cool but those custom carbon fibre bits would scuff badly, (tho' he is a really good rider so maybe he doesn't fall)  and I see Samsung say the 40T is only good for 250 cycles after which capacity reduces to 60% . I think I would find the motor/controller noise annoying as well.

I do like it when others give EUC design a go. He seems to have access to a good carbon fibre fabricator so maybe he should go the whole hog and design his own shells. That would really free up his options. 

[RagingGrandpa]

3 hours ago, Planemo said:

ALL the EUC manufacturers work pretty much the same way (over rating the capacities)

"EUC formula" is:

• Cell Vnom = 3.7V
• Cell capacity = label capacity
• Pack capacity = 3.7 * (label Ah) * (# P) * (# S)

So 24s2p 4,000mAh Samsung 40T yields 710wh.

And the marketing people might round-up to the nearest 100wh, like Leaperkim did

[yoos]

16 minutes ago, RagingGrandpa said:

"EUC formula" is:

While this formula is conveniently universal among manufacturers, making comparison easy as Planemo mentions, it's still useful to know the real energy capacity of the wheel because EUC World report true Wh/km consumption. Knowing your average consumption and real energy capacity you can predict your range quite well.

[RagingGrandpa]

4 hours ago, yoos said:

Knowing your average consumption and real energy capacity you can predict your range quite well.

Did that really work for you?
(Examples appreciated!)

A "real capacity" assumption needs to specify some tricky things, such as cell temperature and pack current.

For example, my best guess at "total electrical output at 1500W average load, before EUC low-battery tiltback" for your Sam40T 24s2p pack would be:

• ~17A/pack, or ~9A/cell
• Tiltback at 3.25V/cell (2020+ Gotway firmware)
• For the 40T cell, Lygte measured 11613mWh/cell for a 10A discharge to 3.2V, with cells starting around room temperature.
  Our load is slightly less, but our end-voltage is slightly earlier, so let's just use their number without a fudge factor.
• 11.613 * 24 * 2 = 557wh delivered.

I think I use about 50wh/mile at 30mph, so range is "11 miles fast" then? (Sounds plausible to me)

 

Edited November 15, 2021 by RagingGrandpa

[yoos]

Minor update: apparently, he reached 65kph on this 20kg ACM. (sorry, no solid proof, just his own comments below his winter ride video). I would love to see a 20kg production wheel with a 55-60kph top speed and this build shows it's absolutely possible (boils down to high-drain batteries).

[supercurio]

It's a good technical achievement, although a bigger tire and heavier weight might still be preferable for stability and handling of unexpected road conditions at speed 

[goatman]

the 40T is one of the best batteries there is right now, going to 2.5v every 100 cycles will regenerate lost capacity back into the cell

you can see my "desulfate" tests of 30q,40t, vtc6 and p42a here

https://endless-sphere.com/forums/viewtopic.php?f=14&t=109601#p1604871

my cycle test of 40T here

https://endless-sphere.com/forums/viewtopic.php?f=14&t=106550#p1560559

cycle test of p42a here

https://endless-sphere.com/forums/viewtopic.php?f=14&t=107573&hilit=p42a+cycle+test#p1574686

the 40T is also the one of the few cells thats like regen and fast charging

 

pajdas test of 40T and others here

https://endless-sphere.com/forums/viewtopic.php?f=14&t=103092&hilit=p42a+cycle+test&start=550#p1675227

 

[supercurio]

Awesome tests @goatman props on the dedication and thoroughness!

Did you have an opportunity to test the Molicel P42A in the end? I wonder how it compares, and how it would perform in a PEV/EUC application.

Edited January 4, 2022 by supercurio

[goatman]

Pajda did it 1000 cycles i think in this thread

https://endless-sphere.com/forums/viewtopic.php?f=14&t=114473#p1692531

 

p42a handles cold temperatures alot better than a 40T

both cells like heat

during the p42a test i couldnt get consistent mah/cycle til i raised the amps near the end of testing but i got all the information i needed for myself and  to settle a disagreement with another member about how "10amp cells" are really just 1xcapacity  cells (2.5-3.5amp cells) and if you want a "10amp" cell for cycle life you need a 20amp cell, no one did those tests and shared the data that i could find

if you chose the wrong cell, packs give out prematurely because of 1 b grade cell that sneaks in and dies

 

the 48x sounded really good til i read the lifecycle test data sheet, it was a very controlled test and i dont ride in a controlled temperature environment didnt want to invest the time testing a "10amp" cell

[supercurio]

Thanks again: awesome reading!

I understand why the 40T is so popular, the performance and durability look fantastic.

[yoos]

In this video he reaches 65kph (that's after correcting for wheel size etc.) while riding on a track (it was neither a race nor a dedicated speed test, just his top speed).

 

[alcatraz]

Why pair the Nik board with a Tesla motor? Is the Tesla motor much lighter than the Nik motor? What about reliability/compatibility?

[yoos]

14 hours ago, alcatraz said:

Is the Tesla motor much lighter than the Nik motor?

I guess so - lighter and smaller (narrower?) perhaps.

 

14 hours ago, alcatraz said:

What about reliability/compatibility?

No idea . Is there much of a difference between current tesla and nik motors? I guess bldc motors are quite simple and typically have enough overhead. Also bear in mind this rider is not particularly heavy or large so he might get away with typically less power demand.