Hub motor on 240v DC ebike? Controller(MOSFETs), windings(phases)...

[nickceouk]

The hub motor is to drive an ebike with none trivial specs.
The idea is that I should be able to drive a hub motor with 240v DC battery (100 lto cells in series or 67 cells log higher density cells 3.6v nominal).
The controller would obviously need to be able to handle 240v DC switching and I have no knowledge of MOSFETs that do or their efficiency.
It strikes me that more than 3 phases may or may not benefit towards my goal.

I also need to check the fundamental assumption here : matching performance of a 48v 1000watt hub motor on a 240v DC battery. 1000watts /240 ~ 5.
Running a hub motor natively at 240v DC at 5amps would yield a similar real world performance to a 48v 1000watt system??

There is another thread I started which was about Street EV charging from CCS DC chargers(public infrastructure)and it's there that the 240v DC battery design started making sense to me.
Keeping a native 240v DC would allow a CCS spec DC charging with minimal hardware at the 240v DC the battery has.
It's a question of working out the controller and hub motor to drive natively at that voltage.

:lol:

 

[MadRhino]

This would be very expansive, custom hub and controller especially. There are simple, cheap solutions to charge existing systems from 240v charging outlets.

 

[nickceouk]

This would be very expansive, custom hub and controller especially. There are simple, cheap solutions to charge existing systems from 240v charging outlets.

Nothing I know of that can charge LTO packs in 10 mins worth mentioning...LTO are low density but can fully charge in 10 mins. The problem becomes more obvious when trying to fit everything on anebike scale setup.

Is the basic idea of maintaining the ~1000 Watts 240v DC at 5 amps workable to achieve a par on a 1000 watt 48v traditional hub motor kit?

The numbers aren't exact and can be made to work with whatever hardware is currently out there - closest to these values.

 

[john61ct]

I believe a re-wording of your query is:

"Are there stock (reasonably compact and not too heavy) ebike motors & controllers designed to work at 57-58S voltages so they can charge at 240Vdc ? "

And afaik the answer is no. Maybe at full motorcycle sizes and weights. . .

I personally think you need to go NMC/NCA rather than LTO, and settle for say 30-40min recharges with the stack of R4850G2's you will now have extra hauling capacity to carry around.

Of course custom builders can make you whatever you like (within the constraints of current technologies, and physics), if you have very deep pockets!

But you're still not folding it up and taking it on public transport, be decades before that's possible.

Just relax and chat with some locals over coffee and a sandwich, why that fixation on a 10-min charge anyway?

 

[markz]

That is lethal voltage levels there, which is alright for a major manufacturer like Tesla, but for hobbyists would be scary.

 

[amberwolf]

there are certainly high voltage controllers, but afaik they're all expensive, and require programming to the specific motor they're to be used with (like sevcon, etc) with hardware and software tht itself is expensive and complex to use. they're made for oem use, and don't generally have any support from teh company that made them for diyers.

there are high voltage diy powerstages for various controller brains, like lebowski and vesc; you'd have to build these yourself.

there's also some projects that use powerstages from large-ev controllers (like the leaf) that can handle high voltages, controlled by the diy brains above.


but none of these are likley to be small or lightweight, as the reason for going high voltage in those cases was to get high power as well. (dozens to hundreds of kw).

they'll still work for your purposes, just that they add complexity to your build and setup process, and likely significant cost and time.


any typical hubmotor will work for you on the high voltage, but will require setting up the controller to more or less have a speed limit, so that the mtoor doesn't get spun up way too fast for your bike, by mapping the throttle to operate the motor only within that speed range, and also to setup acceleration limits so it doesn't just wheelie you right off the bike at the slightest throttle touch.

 

[amberwolf]

a simpler and easier to execute idea is to use a dc-dc converter between the battery and the controller to convert the high voltage down to a regular voltage compatible with the controller and motor you arleady have (or were planning to get).

for instance, you could probably use those meanwell hlg led psus for this, though i haven't tried it, they have been operated by others from dc input instead of ac. if you used the hlg-600h-54a, it can handle 600-700w continuously; about 11-12a or so. if you need more current, use two in parallel. the hlg's ac input wire would be connected to the battery output, and the hlg's dc output wire would be connected to the controller battery input (instead of teh battery).

tehre's other dc-dc converters, out there too, but don't know how reliable they are vs the hlg's. wturber has a thread using a boost converter to change a 36v pack up to a 48v controller. i think he's had one failure but the second is still working.

 

[john61ct]

well he wanted 240Vdc specifically because he didn't want the "extra baggage" of carrying chargers around

exacerbated by using LTO as a chemistry, specifically to get a 10-min refill back to 100%

cross ref https://endless-sphere.com/forums/viewtopic.php?t=104319

with some eBay deals on HLGs in bulk, also Huawei and / or 3K flatpaks

 

[E-HP]

It strikes me that more than 3 phases may or may not benefit towards my goal.

What did you have in mind?

 

[Sunder]

If we're gonna play silly buggers with lethal voltages of electricity, you REALLY gotta work hard to find a novel way to die.

Assumptions - OP has said he wants around 3kwh of battery at 48-72 volts, totaling 3kw, that can be charged in 10 minutes, without massive chargers, and using off the shelf motors and controllers. Impossible? Only if you have no imagination.

Step 1. Build a 25S5P LTO battery - 60v nominal, 67.5v peak charge. So you're looking at 10Ah cells x 5, or a 60v 50Ah battery. Controllers and motors that handle 60v nominal are a dime a dozen. Easy

Step 2. Build a bridge rectifier. 240vac * 1.41 = 339vdc, or there abouts.

Step 3. Reconfigure 25S5P into 125S1P. This is now a 337.5v. 10Ah battery fully charged. Are you picking up what I'm putting down?

Step 4. Calculate or measure approximate internal resistance of battery, and select two resistors - One that will allow you to charge at 10A (1 hour charge), so you can use a normal house hold power point, and a second that will allow you to charge at 60A, for an EVSE that will support it. This does introduce complexities, as your rate of charge will fall as the difference in voltage shrinks. But do your maths right, and you'll get close.

Step 5. When the charge is finished, reassemble your battery to a 25S5P battery to put back into the bike.

Step 6. Actually don't do that. I don't want the BBC to be reporting a torched corpse at one of the new Street EV chargers, and demanding an investigation as to why they aren't safe.

Seriously though, I love LTO. Everything I have is LTO. I have a 60S, 30AH battery for a motorcycle, and a 20S 4Ah battery for an electric bike. Yes, my LTO can theoretically charge in 10 minutes, but the hassle of charging it that fast outweighs the specialist equipment and unreliability of finding charging point that can do it. If I limit my charger to 2400w, I can go to any cafe, car park, hell if I got in a pinch - any stranger's house, and ask to charge. You go any over that, and you're 100% reliant on finding a EVSE compatible and rated to your charger (And you know that many chargers share current, so even if it can deliver your full current when you get there alone - as soon as a second car plugs into the same circuit, it sends a signal to the car to renegotiate the charging rate).

It's a dream, but these batteries were mass produced for Chinese EV buses, with expensive and specialist infrastructure just for buses. We wanted them 7 or 8 years ago when Toshiba first released them to OEMers only, and they were unavailable to DIYers. It wasn't until China made them in bulk that they became available (but also why they are almost all 66160 in size, or duplicates of the original SCiBs.) And without that OEM budget and infrastructure backing you - The 6C charging will be just a spec on paper you can't realise into a practical, workable solution.

 

[dogman dan]

Re charging in the wild, unfortunately my town pretty much locked up all the plugs in city parks and such, preventing no hassle charging on the go.

It definitely makes sense to be capable of using a tesla charge station if you have them, using some kind of converter down to a reasonable bike voltage, like 72v. Then give it the fastest rate of charge your particular cells can easily handle.

Make the whole thing work with car voltage? Sure,, on a larger motorcycle, that costs a fortune.

 

[nickceouk]

It strikes me that more than 3 phases may or may not benefit towards my goal.

What did you have in mind?

This was a directed at hub motor design.
No specifics unfortunately

 

[nickceouk]

If we're gonna play silly buggers with lethal voltages of electricity, you REALLY gotta work hard to find a novel way to die.

Assumptions - OP has said he wants around 3kwh of battery at 48-72 volts, totaling 3kw, that can be charged in 10 minutes, without massive chargers, and using off the shelf motors and controllers. Impossible? Only if you have no imagination.

Step 1. Build a 25S5P LTO battery - 60v nominal, 67.5v peak charge. So you're looking at 10Ah cells x 5, or a 60v 50Ah battery. Controllers and motors that handle 60v nominal are a dime a dozen. Easy

Step 2. Build a bridge rectifier. 240vac * 1.41 = 339vdc, or there abouts.

Step 3. Reconfigure 25S5P into 125S1P. This is now a 337.5v. 10Ah battery fully charged. Are you picking up what I'm putting down?

Step 4. Calculate or measure approximate internal resistance of battery, and select two resistors - One that will allow you to charge at 10A (1 hour charge), so you can use a normal house hold power point, and a second that will allow you to charge at 60A, for an EVSE that will support it. This does introduce complexities, as your rate of charge will fall as the difference in voltage shrinks. But do your maths right, and you'll get close.

Step 5. When the charge is finished, reassemble your battery to a 25S5P battery to put back into the bike.

Step 6. Actually don't do that. I don't want the BBC to be reporting a torched corpse at one of the new Street EV chargers, and demanding an investigation as to why they aren't safe.

Seriously though, I love LTO. Everything I have is LTO. I have a 60S, 30AH battery for a motorcycle, and a 20S 4Ah battery for an electric bike. Yes, my LTO can theoretically charge in 10 minutes, but the hassle of charging it that fast outweighs the specialist equipment and unreliability of finding charging point that can do it. If I limit my charger to 2400w, I can go to any cafe, car park, hell if I got in a pinch - any stranger's house, and ask to charge. You go any over that, and you're 100% reliant on finding a EVSE compatible and rated to your charger (And you know that many chargers share current, so even if it can deliver your full current when you get there alone - as soon as a second car plugs into the same circuit, it sends a signal to the car to renegotiate the charging rate).

It's a dream, but these batteries were mass produced for Chinese EV buses, with expensive and specialist infrastructure just for buses. We wanted them 7 or 8 years ago when Toshiba first released them to OEMers only, and they were unavailable to DIYers. It wasn't until China made them in bulk that they became available (but also why they are almost all 66160 in size, or duplicates of the original SCiBs.) And without that OEM budget and infrastructure backing you - The 6C charging will be just a spec on paper you can't realise into a practical, workable solution.

I should mention the other LTO thread that started this discussion https://endless-sphere.com/forums/viewtopic.php?f=14&t=104319
The series to charge parallel to hub motor idea is one of 3 options that emerged for me on that thread.
Thanks for bringing the specifics and yeah safety comes out top in any possibility I wish to explore.

I hope you follow your own advice to be honest - practice what you preach- drop the LTO it's got no workable benefits in 2020 :confused
You know 60s30ah - I don't want the BBC.... :

About the only and primary benefit of 10min charge is also working against the LTO super safe chemistry when factoring the practicalities of hardware needed to feed all that power through the cells at those voltages, amps and bulk.

So yeah - overall no benefits to achieve ebike touring on LTO cells and hence not a worthy route for me.

 

[nickceouk]

Re charging in the wild, unfortunately my town pretty much locked up all the plugs in city parks and such, preventing no hassle charging on the go.

It definitely makes sense to be capable of using a tesla charge station if you have them, using some kind of converter down to a reasonable bike voltage, like 72v. Then give it the fastest rate of charge your particular cells can easily handle.

Make the whole thing work with car voltage? Sure,, on a larger motorcycle, that costs a fortune.

I am down to 2-3kw charging ball park - NMC/NCA route.
Street EV charging is solid and a matter of time - who wants to wait through

There are Type 1(3kw AC) and Type 2(7kw AC) chargers with good enough coverage to be worth tapping into for ebiking.
There is also the gorilla charging route(cafes, restaurants, fast food and casual 'neighbours')
which is always good as a backup.

P.S. it's possible for a crowdfunding product to tap into Tesla's and other DC chargers to serve ebikes. Then again it's just one of many many possibilities as well , right?

 

[john61ct]

Believe me soon as any real customers start getting inconvenienced by guerilla DIYers, those subsidising the EVSEs will block our using them.

 

[Sunder]

I hope you follow your own advice to be honest - practice what you preach- drop the LTO it's got no workable benefits in 2020 :confused
You know 60s30ah - I don't want the BBC.... :

About the only and primary benefit of 10min charge is also working against the LTO super safe chemistry when factoring the practicalities of hardware needed to feed all that power through the cells at those voltages, amps and bulk.

So yeah - overall no benefits to achieve ebike touring on LTO cells and hence not a worthy route for me.

Fair cop. A few years ago, I did brush a soldering iron across about 80-100v of the 36S (LiPo) battery in my signature. Flung molten metal everywhere, including burning holes in the shirt I was wearing. Amazingly, nothing hit my glasses, and even the bits that burned through my shirt, didn't touch my skin. After that, I taped down every single wire before soldering any further up the battery. That 60S battery was also assembled as 3 x 20S batteries, and the inter-connects assembled with no load on the final terminals so they wouldn't arc. I also 3D printed myself a plastic pair of pliers, bought plastic screws and a set of insulated screw drivers to work near it. (Tried 3D printing a screw driver. Even ABS/PETG wouldn't screw in a plastic screw into a pilot hole without snapping).

But perhaps that, and a couple other close calls is part of the reason I'm a bit sarcastic at people taking high risks, unless they're someone like the blue named "Guru's" here. I also generally don't play around with mains level power. I use something professionally made to step it down before use.

When you say "Overall", I'd still disagree with that. There are many benefits, but 6C charging, is one I don't think we can claim. Here's why I use it despite the cost, weight and volume:

1. Once assembled, you can abuse the crap out of them. Low voltage cut off? I've run some down to zero and left them that way for a month. No puffing and ZERO loss of capacity of power density. My dad has a set in his farm shed, feeding the house. It gets to 65-70*C in summer there. Been going strong for 3 years. During a flood, the bottom row got to zero, while the top row got to 5+v (same event as above) - no fire, and no damage to top cells either.

2. Just because you can't get 6C for 3kwh, doesn't mean you can't get 2C for 400wh That's only 800w. That's enough for my commute eBike in just over 30 mins (Charging always slows when you get near full, so 80% in about 20 mins, and the rest is pretty optional.

3. Cold weather charging. Less of an issue in Sydney, as we get maybe 5 days under 10*C, but it only takes a few cycles of charging LiPo/Li-Ion even moderately quickly (as little as 0.3C), to irreversibly lower the power density.

4. Safety. You can throw these into a fire, and they won't burn. Of course, anything over about 50ish volts can still start an electrical fire if you put it near something else that burns, but the batteries themselves won't burn.

 

[john61ct]

And the self-heating aspect, risk of thermal runaway so much lower.

Even LFP can get dicey if charging controller fails, LTO just a whole nother order of magnitude safer in that regard.

 

[markz]

Very crazy for ebikes, just to charge at ev station? You might as well buy a motorcycle, gas or electric dont matter.
Charging a bicycle at a EV charging parking stall will raise too many eyebrows.

 

[dogman dan]

I'm saying get a motorcycle more and more, and not entirely because an e bike burned my house down.

Its just crazy cheaper for me to ride freeway speed on my 400 cc scooter, than electric. And I'm insured doing it which might be good.

Others have more downside to this,, gas here right now is stupidly cheap, and insuring a 60 year old on a motorcycle is too. But the cost per mile of my 40 mph e bikes has been nearly as much as car driving, while the scooter is barely over a dime a mile. It never needs batteries every two years, just a sip of that cheap gas.

Still love my e bikes, even after the fire. All batteries stay in an outdoor bunker now. But I don't try to make home made motorcycles anymore, the gas ones are just too cheap on the used market here.

 

[E-HP]

It strikes me that more than 3 phases may or may not benefit towards my goal.

What did you have in mind?

This was a directed at hub motor design.
No specifics unfortunately

Interesting. Can you post a link with more info on a hub motor with more than 3 phases? Curious about how that works.

 

[nickceouk]

It strikes me that more than 3 phases may or may not benefit towards my goal.

What did you have in mind?

This was a directed at hub motor design.
No specifics unfortunately

Interesting. Can you post a link with more info on a hub motor with more than 3 phases? Curious about how that works.

There is someone on the other thread(before I started this one)and he claims 6 phase motor with some other eyebrow stuff to match
https://endless-sphere.com/forums/viewtopic.php?f=14&t=104319&start=25#p1523855

I found a page while waiting and regarding more than 3 phases they were adamant it's not worth the effort. I think it was this discussion towards the bottomhttps://electronics.stackexchange.c...phase-power-why-not-a-higher-number-of-phases

My intuition tells me there is a whole lota stuff to tap into. Some of that belief and interest and is based on those "Ferocell" YouTube videos
[youtube]https://youtu.be/c1V4I_UZtyE[/youtube]
. Fringe science is probably accurate :lol: There is enough there to raise some "powerful and valid questions" if your mind is geared for it. Everything else is on the other side of speculation so I could be wrong :wink:

 

[amberwolf]

there's a number of discussions on es about non-3-phase motors, but i couldn't find anything that covers them all. most of them start about the falco motor or the tidal force motor, and bring up others in the discussion.

there's also a few about 6 phase vs 3 phase.

some of them are in this search, but most of the list isnt' relevant.

https://endless-sphere.com/forums/search.php?keywords=*phase*+motor*&terms=all&author=&sc=1&sf=titleonly&sr=topics&sk=t&sd=d&st=0&ch=300&t=0&submit=Search