Ebike battery

[Hopup]

Inspired from some videos from internet, I was thinking making long range ebike as much diy as possible.

I don't have too good understanding of batteries in this type of application, but would it be feasible and ok to use 18650 cells to build such a battery bank for the bike?

I am worried about charging it and its safety I would want to have 1-2Kwh size bank.

 

[Fish4Fun]

Hopup,

If you are thinking of the cheap 18650 cells from ebay you will almost certainly be disappointed. There are a number of things you need to take into consideration With-Respect-To (wrt) battery packs in the KwHr+ domain.

First you need to make 100% certain the battery array is capable of providing the requisite current. I would think your ebike motor would probably need at least 20A and quite likely upwards of 50A in short bursts. There are plenty of 500wH+ battery packs designed for < 5A discharge rates ... examples include various camcorder battery packs and many of the 18650 type batteries. In order to get 50A from 5A rated battereis, you would require 10 of the 5A batteries in parallel.

Lithium batteries range from ~3.5V to 3.9V per cell depending on type, with 3.7V being fairly common. There are numerous motors designed for ebikes and scooters, ranging from ~12Vdc to ~60Vdc in both BLDC and brushed configurations. The advantage to higher voltage is that the i^2 * R loses are significantly reduced for any given power rating ... for instance a 600W motor would require 50A @ 12V,dc but only 10A @ 60Vdc. Assuming the motor leads have 0.1 ohms of resistance, the losses in the leads alone in the 12V version would be 50^2 * 0.1 = 250W. The same leads to a 60V motor would only lose 10^2 * 0.1 = 10W.

Assuming a nominal motor voltage of 60V, this would imply roughly 16 * 3.7V batteries in series ... Assuming each of these batteries is 3Ah, this would provide you with 180wH .... 1kWh would require roughly 6 of these arrays, 16 * 6 = 96 batteries 3.7V * 16 * 6 * 3Ah = 1.065kWh. For 2kWh obviously you would need twice as many batteries.

My advice would be to use one of the battery packs // motors // speed controllers designed specifically for ebike conversions and don't try to DIY more than the assembly.

Good Luck!

Fish

 

[Hopup]

Thanks for your answer. I was thinking of using center motor and higher voltages for the better efficiency. Probably around 500-1000w motor. Batteries either from Aliexpress or quality brands. Something around 50 to 100 should be enough.

It is certainly better to use high voltage in this application. Those aliexpress batteries should still provide at least 2 amperes so you would just need more of them. It would be better to go with higher quality.

It would obviously be better to have complete kit but that would take whole diy aspect out of it. Plus I think the savings can be big.

 

[dave9]

If you reap any cost savings, you'll end up paying for it 10X over in time and frustration, or suffer longevity in lifespan or travel distance.

Merely adding up # of cells needed in a battery, is nothing compared to assembling a pack, using a spot welder, into an appropriate size fire-containing metal enclosure, custom charger, battery health monitor, motor controller, etc. You'll be DIY what others have spend cumulative lifetimes worth of engineering and experimentation to achieve. HOWEVER you can leverage that, if you frequent an eBike forum so you can copy some existing designs and ask questions of those who build these as an obsession, but they will probably tell you the same thing, to buy ready made components until you know of reasons to deviate and improve on those.

I'd never think in terms of "just need more of them" when choosing cells. The inexpensive ones aren't just low current capacity, they're typically fraudulently overrated so at the end of the day, you aren't really getting much if any more capacity per dollar, more so that it's just building a bigger, heavier battery.

However, it's really easy to find batteries capable of more than 2A current. One option "might' be harvesting them from new cordless tool battery packs. You can sometimes find these battery packs deeply discounted, for example when a tool manufacturer switches over to a different battery format and closes them out, but I'd make sure to choose a major brand or one known to at least have major brand cells in it, often determined from a youtube battery teardown video, not using those 3rd party generic battery packs which are practically always shown to have poor capacity relative to their specs.

It is a bit beside the point, that this is not the right project to learn about Li-Ion battery packs, particularly on a vehicle where you are traveling and located so close to the battery. It could become quite the fire hazard if 100 cells were to cascade fail, then you are stranded, fire department comes and puts out your bike fire then a wrecker hauls it away and bills you for that service.

If you want a reasonable price point for travel or just more fun than burden, I'd sooner recommend a gas engine scooter or motorcycle.

 

[Hopup]

I don't think the system needs to be very complex in electric bike, after all it is just electric motor, controller and battery pretty much.

But yes I am too worried about the battery maintenance. I do have spot welder so building is not really a problem. It would just need good circuit for charging and balancing. I would most likely buy that part I don't think I could make it.

Electric scooter or motorcycle is not really same, you cant use it when you run out of battery, and you cant use it outside of roads legally.

Maybe smaller battery would suffice if efficiency is good maybe 0.5kwh since that would get around 60-80km potentially.

 

[dave9]

^ Underestimating the complexity, is why you're likely to have a fire, and personal injury. I can say a car is "just an engine, wheels, and a cabin" but we both know the devil is in the details.

I did not mean an electric scooter or motorcycle, the exact opposite, a used gasoline engine scooter or motorcycle. They *can be* very fuel efficient, so you are paying a few pennies per mile instead of an up front expenditure of several hundred dollars in parts, and several hundred hours in development and testing.

Why would you need to use it outside of legal roads, if it goes fast enough to travel on roads? That seems the superior method of transportation for any vehicle that can manage it legally... kinda why roads exist.

Regardless, an eBike forum is where you really want to be, is set up for exactly this purpose. They will probably be able to tell you where to buy cells to build a battery in the most cost effective way possible to reach your target travel range, especially in whatever country you happen to be located in. I don't think that Aliexpress is going to be the way to go here, unless you want a high likelihood of counterfeit batteries or generics that are grossly overrated. When you deal with a merchant located in your own country, you have more legal leverage to get product issues resolved, especially if trying to do a bulk order for hundreds of dollars worth of cells.

 

[Hopup]

Yes I agree need to find some ebike forum they can probably recommend much better than in general place like this. Yes I know it is more complex than just couple parts but I also know that plenty of things can be usually simplified without losing the main feature (electric bike propulsion).

I know you mean gasoline bike, but if I wanna cruise with the bike in smaller roads it wont really do just because of its legality in where I live required gasoline engine vehicles to travel on road where cars do. The kits are like 700-800euros for good one which is a lot for what you get in it I am sure simple kit can be diy cheaper I dont see anything too complicated if we only focus on the main feature of powering the electric motor.

 

[Audioguru]

AliExpress sell fake 18650 batteries like the other Chinese places with "sellers".

 

[Hopup]

AliExpress sell fake 18650 batteries like the other Chinese places with "sellers".

I don't think all are fake. Probably "faulty" batteries discarded by the manufacturer. Low amps.

 

[Audioguru]

YouTube has videos of cheap ebay Lithium batteries taken apart and showing flour inside instead of battery chemicals.
A battery advertised by AliExpress shows its impossible 9900mAh rating. Look at its very low price.

 

[Bluejets]

I wanna

Dead give away here.

 

[Hopup]

YouTube has videos of cheap ebay Lithium batteries taken apart and showing flour inside instead of battery chemicals.
A battery advertised by AliExpress shows its impossible 9900mAh rating. Look at its very low price.

Yes, obviously there's some fakes there.. 0.39 is obviously fake. You have to buy products based on reviews on aliexpress and similar sites. I have bought plenty and never got bad product.

 

[Hopup]

Dead give away here.

About what? That's how people talk nowdays, at least on the interwebs

 

[Audioguru]

Reviews are also easily faked.

 

[Hopup]

Sure but that's not the case at least for the products I have seen. Some common sense, little bit of intuition and you can buy decent stuff from Aliexpress. Cant talk for other sites tho, got some bad stuff other places, well the quality was not very good, yet no reviews either.

 

[Fish4Fun]

@Hopup,

I think what people here are trying to tell you is that in today's economy many things are cheaper in both cost and labor to purchase as an end-product or a "kit" specifically designed for conversion of an existing product than to attempt to engineer and build from scratch.

Take for example the Arduino Mega2560 boards. They range in price from ~$7 to $15 delivered to your house with a USB cable from places like ebay. If you wanted to "DIY" a similar project, the ATMEGA2560 alone would cost you ~$12 + shipping ( https://www.digikey.com/product-det...ogy/ATMEGA2560-16AU/ATMEGA2560-16AU-ND/735455 ) the USB interface IC another couple of dollars, the headers, crystals, capacitors, LEDs, voltage regulators, etc, etc could easily cost you another $20 by the time you rounded them all up and got them delivered to your home. Next you would need to draw up a schematic and then layout the PCB .... Finally, you would need to export the Gerber files and send them to a PCB mfg. A few days to a few weeks later you would receive your PCBs in the mail. Assuming you did everything correctly up to this point, you could then begin the task of attempting to solder the 100-TQFP ATMEGA into place. After destroying a half dozen boards and chips you might finally get one of the 100-TQFP ATMEGAs soldered in place properly. Once that chip is soldered you should be able to solder the rest of the components in an hour or two. The final step would be to download the USB library for the CH340 USB interface, compile it and then flash it. Assuming everything went smoothly, you would have ~$250 + 100 hours in your first finished board .... Even if you needed 5000 of them, it would still be cheaper AND MUCH FASTER to buy them already made than attempt to "DIY" them.

Like your "ebike", the MEGA2560 boards aren't anything complicated ... honestly they are very straight forward, but that doesn't mean building one from scratch is cheaper OR easier than simply buying one.

Have Fun!

Fish

 

[Hopup]

Yes, Fish4Fun I agree on the example of Arduino. Yet making electronics product like that is quite a lot more difficult than making cruder version of electric bicycle at least if we don't aim for perfection.

I do suppose that it is not really worth the price to do ebike unless I will also use repurposed motor from some used appliance and use some cheap controller module. But in that case + keeping the system as simple as possible (no regenerative braking or fancy modes) it would be not only considerably cheaper than lets say 1.5k to 2k electric bicycle but also more dangerous hah.

But anyways point noted and I will be looking ebike forums if I will decide to move on with that project which is not certain at this point.

 

[Bluejets]

About what? That's how people talk nowdays, at least on the interwebs

Whatever....that's also how I hear it sometimes, still doesn't change the fact.

 

[Fish4Fun]

@Hopup ...

I am NOT trying to discourage you from retro-fitting a bike to make it an ebike ... I am simply suggesting this process, like the ATMEGA boards, is "mature" ... thousands and thousands of hours have been dedicated to engineering and testing various solutions to achieve the finely tuned, low-cost solutions that appear as readily available, (relatively) inexpensive "kits". I would give you the same advice about attempting to build a bike from DIY parts in an effort to save money over simply buying a fully assembled bike.... Even if you needed to replace various parts on the "stock bike" to make it suit your needs, it is still generally much cheaper to replace a few stock parts with custom parts than it is to build an entire bike from "custom parts".

View the ebike conversion kits the same way ... they simply give you a pre-engineered solution for a fixed price and offer you the best chance of completing your project within an 'a priori' budget.

I am working on a project right now using a motor marketed for e-bikes and e-scooters ... ( https://www.ebay.com/itm/4600W-6384...e=STRK:MEBIDX:IT&_trksid=p2057872.m2749.l2649 ) I have already dedicated 100+ hours and well over $100 in components designing, building and testing a driver for it in the FULL KNOWLEDGE I could simply order one of these ( https://www.ebay.com/itm/48V-1500W-...938464?hash=item568cf33ca0:g:aqAAAOSwGj9dz65k ) for $34. Obviously I have specific criteria for my driver and saving money is NOT a priority; but I still fully investigated existing drivers before making the decision to design my own.

Going back to your original post, even if you decide to DIY everything else, I would strongly urge you to consider buying a battery pack specifically designed for e-bikes. I cannot see any way that you could save money attempting to DIY a battery pack, and the chances of catastrophic failure in a DIY battery pack are very high. To be clear, by "catastrophic failure" I mean FIRE with the possibility of EXPLOSION.

For an example of exactly how badly this could go, https://www.schmidtlaw.com/dell-laptop-fire-lawsuit/. While a lot of the story has been buried, the base story is that, against advice from Sony, Dell decided to save a little money on their laptop batteries ...

Good Luck!

Fish

 

[Hopup]

Good post. Is that motor any good for ebike, what is its real power capability? 100 dollars in component design is not so bad if you are looking for improved design which you seem to have been doing. I prefer simplicity over more features.

Yeah it seems like battery is kinda dangerous, but I have no doubt that I could not assemble the pack properly. More about charging it, which can be solved with pro charger. If you keep the cells in balance, and have fuse and some form of control for the temperature I don't see it exploding with me in it. But yeah it would be considerably wiser to buy pack (I guess).