Ebike build project

Hello from Greece. My name is Vaggelis and i’ve been reading this forum for a while and i can say that you guys are very helpfull. I’m building an ebike ( like Tom Stantons project) and so far i have the bike ( single speed) and the motor ( keda 63-64 190kv). I need an esc and i see most people here like the focbox from enertion. The problem is, that it is a little over my price range and while i was looking for a cheaper alternative ( if there is one), i saw one on hobbyking esc at 89$. Some of you might say that it’s not a big price difference, but does it worth it? Because i also have to make my battery pack ( li-ion) and get a balance charger ( or ad a bms and use a regular charger) and i can’t go cheap on those. Any help would be appreciative, cheers.

If you’re building an e-bike, you will draw more current. That esc isnt know to be good quality, Because it is a 4.10 hw version. If you want something reaible Get a brand new focbox from @Titoxd10001 It cost $125 plus shipping, and is definitely worth it

Is he from US? I can only get it from EU, because of custom taxes.

He made it and may be can help @wafflejock

He made an ebike you mean?

Maybe try asking here, there are much more ebike builders who might help you with your project :slight_smile: Also much more topics more related to your build :wink:

I already did, but then I found this forum and it looks like you know your staff.

Endless loop really is the place for ebike stuff. I’ve experimented with moving my esk8 parts including one of the older 4.10 hardware VESCs over to a bike (using Stanton’s parts for the motor mount and wheel pulley), but so far mixed success. I haven’t run into issues with burning out the VESC yet (on a bike), but I have followed Tom’s stuff for a while here and he did mention he blew at least a few VESCs and did eventually start using FOCBox (maybe it was vesc6 I forgot but pretty sure he got upgraded for free).

I printed and molded the parts for the wheel pulley but unfortunately the tooth spacing doesn’t seem to be 100% correct. Same issue is present on the original 3d print as on the molds but I didn’t have a belt to check when I did the 3d print and assumed it would be good, so if you’re doing the 1/4 pieces would verify your belt fits snugly before proceeding to make the whole thing.

Aside from tooth spacing I ran into a couple of other issues (possible exacerbated by the imperfect tooth spacing). One of the other issues was if the pulley on the wheel isn’t 100% concentric with the wheel itself then there is some play as it rotates the belt gets tighter and looser (obvious in hindsight but didn’t really consider this as a major problem). To get around that one I’m using an idler on a spring load to keep some tension on the belt but allow it to move to compensate for the lack of perfectly concentric wheel/pulley. Last problem I ran into is needing to adjust the motor mount so that it would be able to be mounted on the particular bike I was putting it on (craigslist garbage mongoose bike for $45 to experiment with). Actually the last bit about the bike brings up another point, make sure the bike has a big enough wheel diameter so the motor doesn’t end up in the way of the wheel itself, I tried on a foldable cruiser someone gave me to try out but with smaller wheels the motor position ended up right next to the wheel (you want regular mountainbike or street bike size wheels, also street bike wheels are ideal, I think in terms of the bike a hard tail 1-by is ideal, see seth’s bike hacks on youtube for some good beginner tips on picking bikes in general etc.)

^ there’s a video testing my setup upside down, I was able to do a little bit of testing riding around before Chicago entirely froze over, but have been working on making my own skateboard wheel pulley design as an intermediary step to redoing the bike one (ideally so I can print it all in one go) but need to do the math etc. on gearing to see how that will work out… basically how much torque will I be sacrificing if I make it small enough to print on a 200x200mm bed.

Custom Pulleys

I’d say if you’re just trying to get something working then going with a hub motor setup is a ton easier (avoid all the alignment issues and potential for belt slipping off in different directions and trying to avoid having spokes run into any of the motor mount parts, long story short alignment is more difficult than you think).

1 Like

Thanks for the tips! I know where i’m getting into though. I know the problems and the difficulties and i think i will solve them ( with a lot of help). My base is an old single bike ( restored) and right now i’m trying to fit a freewheel on the left side and then bolt the pulley on the freewheel, so when i lay of the “gas”, the wheel will keep spinning without draging the motor. But, until i solve that issue, i have to find a good esc.

Yeah so long as you limit things properly to avoid going over 60,000erpm (8571 motor RPM on a 14 pole or 7 pole pair motor which is typical) you should be fine. So you take that 8571motor rpm/190kv = 45.1V.

A 10S battery is 42V fully charged a 12 S adds another 8.4V fully charged so 50.4V (10S would be safe 12S would be too much)

Above that 60,000erpm there is a high probability that the DRV chip on the older VESC will blow, it happens pretty consistently so best to avoid that if you go older hardware. Also FOC mode is known to cause issues on the older hardware so I wouldn’t risk it, just run BLDC, if you can get a sensored motor I’ve been told it helps a lot (makes sense the motor controller can actually tell the position of the motor without relying on back electromotive force to infer the position).

If you keep the stuff above in mind I don’t think you’re likely to blow the VESC even using the old one but newer ones have improvements to mitigate some of these problems (FOCBox gets past issues with FOC mode and VESC6 gets passed the erpm issues as well)

1 Like

Also regarding costs on components I think you are better off buying balance chargers than getting a cheap BMS that just ends up causing you problems. You can also use LiPos instead of 18650 based to bring your cost down a lot. Getting a cheap 18650 cell is IMO more dangerous than getting a cheap LiPo pack, you need cells that can deal with the amperage being pulled through them without overheating and damaging themselves. If you need to go cheap just go LiPo and balance charger, if you have an extra few to five hundred to put into the project then 18650 and BMS.

I haven’t thought about batteries yet, but i understand what you mean. I already have the motor ( keda 63-64 190kv). If i go lipo, i’ll have to use two 5s in series, but i’ll have to charge them in parallel with a balance charger, is that correct? The good thing is, that i have a couple of people here to help me with the electric parts ( batteries, connections etc.).

Yup you would hook them in series for getting 10S and you could charge them in parallel. I find that I’m actually limited by the charger I have so I wouldn’t benefit from parallel charging so I just do them back to back. Since I have an 80W charger if you do 80W/ (5S*4.2V) = 3.8A in reality I can only charge at 3.6A (charger gets toasty) which is still under 1C charge rate for 5Ah 5S batteries. So if I were to parallel charge them it wouldn’t go any faster I just wouldn’t have to swap batteries between charging, but I don’t really mind, I designed my battery tray so I can slide them out to make that not such a PITA (at first just had balance leads hanging out of the box, also works but not as safe/convenient).

swappable batteries

Battery box link below if you’d like to tweak these, onshape is free for public models and cross platform so I use it: https://cad.onshape.com/documents/15cd342c90ae1b0b6fd666bf/w/40780c7d8d67fd69a32fbf6e/e/aa247b01c3dd60bc5ba24701

VESC and receiver and metr/bluetooth/telemetry module box below https://cad.onshape.com/documents/d8179a83484264c79e6adb1a/w/af2067b944ad3aa57f532370/e/35b4562d3dcad3a56c9b531f

Also to note charging at less than 1C is better really, slower is generally better on the battery health but in terms of waiting for it to fill the 5000mAh it’s not ideal unless you are testing your patience :slight_smile:

Ok, so my thoughts are correct ( i’m a noob). I don’t mind charging batteries back to back. This is for an ebike and my route is only 4.5klm, slightly uphill, so on the way back the mototr will be off. Thank you!

No problem use the magnifying lense/search in the top of the page to find stuff on different battery setup options when you get to it, there are tons of threads covering different options/setups.

Also on esk8 roughly 10Wh = 1km so 4.5km should be no problem with a relatively small battery.

Thanks @wafflejock, right now my concern is the esc. I want one to use plag and play if it is possible. I don’t mind configuration, but since i’m noob to this, i don’t want to brick it.

Gotcha yeah was just getting into batteries since that is one way to brick it :slight_smile: too much voltage combined with too high kv motor ends up with too high erpm which blows the chip that deals with getting the motor to turn at that speed.

Reason people all go with the VESC is it’s pretty reliable for “high voltage” applications whereas there are not a lot of other RC consumer ESCs that can handle the voltage (they are made for at most big RC cars like 1:8 scale). Also the ability to configure it means you can make it work for different applications and can do things like reduce the max motor amperage which in turn reduces the torque/acceleration, and you can limit the low voltage cut off with the VESC itself so it reduces power to the motor as the batteries are getting low and once they are too low it cuts off power entirely to avoid damaging the batteries (called low voltage cut off, some other ESCs offer this, often configuring them involves doing voodoo with a controller or using a proprietary windows only app).

If you go lower voltage like 6S then can use X-Car Beast, or FVT 120A with active cooling on them (fan on heatsink) but I’m not sure they have higher voltage versions. I tried a few other ESCs but you want one that has variable braking (and ideally regenerative to get that power back).

I also blew a couple other ESCs, believe one was made for boats and the other was made for planes, believe the plane one had “brakes” but it was just instant braking to stop a propeller not something you really want on a skateboard or bike.

I’m still trying to figure out which one is compatible with my motor, max current 90A. Sure, when you can configure it to your needs, it’s great, but also pricy. So instead on spending money on cheaper, but not reliable escs, it’s better to invest on a vesc. Any way, when this project evolves, i will definetely need your help. Cheers, Vagos.

1 Like

Sure once you have a better idea of what components you’re thinking of putting together can just put together a "build post’ showing what the plan is before you buy things then can avoid returning or spending twice on parts if I or someone else sees something that isn’t going to work.

In general I’d say most people don’t hit the max limits of their hardware anyhow but if you want to figure out what you need to do to allow you to use the motor to it’s full potential you would take the motor wattage and divide by your battery voltage to get the amperage that would need to get to the motor to max it out, that said I have a 2200W motor and I push about 20A @ 40V or 800W to it at most when accelerating hard (but I’m lightweight and don’t ride crazy and live in Chicago).

Running higher voltage means less amperage to get the same watts to the motor, this is the reason we want to run higher voltage when possible too just to clarify.

Ok, i just realized something…i have a lot of reading to do. Thank you, again. I think this is my “build post”.