Skike - A eSK8 parts based eBike

I was looking at AliExpress again, there are a decent selection of gears for around 10$ a piece, if I find this lacks torque, going 10S and gearing down more is definitely possible

1 Like

Going for the first almost completed test drive

One P group is higher than the rest and this shitty BMS doesn’t have balance, I shouldn’t even bothered to install it

1 Like

5 km ride and 4.9 Wh/km, this should get me 40 km of assistance, not having to constantly hold the button is a bless, and having three power levels makes for a better matching between terrain and assist, most of times the full 35 A is too much and a waste of battery

In the end I went full power without pedaling to see how the temps held up, almost 80 °C on both motor and FSESC, I guess during the day and on longer rides it will thermal throttle, the motor cools really fast

2 Likes

Another ride, 3.2 Wh/km :grin:. Potential to 60 km rides

6 Likes

Off topic but make sure you don’t cut your finger on that phone screen, looks like the digitizer pad is lifting a little.

2 Likes

It’s fine :slight_smile:

What impress me most is that this iPod must be almost 5 years old or more, and the battery still lasts a long time

And it sat fully discharged in a drawer for more than 2 years

3 Likes

Batteries are magic :grinning:

1 Like

Discovered a fatal flaw, it you ride on the right (wrong) type of mud, it starts building up on the motor, and since the motor is hot, it dry fast and become really hard, up to a point everything starts vibrating, if not cleaned regularly I fear it might snap the motor or pop the tire

Some parts I was having to stop every 500 m

2 Likes

You could attach a scraping thingy on the mount?

3 Likes

Good idea, will see how I can mount it

1 Like

Also that motor isn’t exactly sealed, be mindful of rocks and whatnot making it into the stator and scratching or cracking everything. I use thin metal mesh that I can bend around contours.

2 Likes

I’m keeping an eye on it, this motor already have a lot of miles on it from my previous boards, and still going strong

2 Likes

Carry on then brother.

1 Like

The bike is doing great apart from the minor problems already described

A friend of mine is doing one, but with my initial idea of two motors with a two or three stage gearing, one that can climb everything. Given that, I added more functionality to my spreadsheet, now it can calculate acceleration time for various speed interval and grades

Unfortunately there is a nasty bug that is screwing with the calculations when motor and battery is not the same, need to dig a bit more on it, and also translated it so anyone can use, once I fix it will post here

here is just playing with the idea of a electric motorcycle using a bunch 63XX motors

3 Likes

Sorry to backtrack - why couldn’t the sprocket also be the brake disc? Just make a water/laser-cut brake disc with sprocket teeth. Of course the caliber would need to be mounted on the chain stay, or very low on the seat stay. Just need to make sure not to lube the chain!

If you go down this path check out Rocky Mountain’s mid drive design. The chain torque sensor is positioned to measure rider input regardless of what the motor is doing. This means they don’t need all the bottom bracket and cadence sensor BS like Shimano Steps, etc. It’s where I got the idea to make the simple resistor based sensor I posted earlier.

image
Also this system shouldn’t need a freewheeling crankset (unless you get a flat battery) Most mid drives need this feature to hide motor lag. Either you’re slow to push the button, or if it’s a traditional (BB) torque sensing unit the controller has to wait for cadence sensor input to be sure you aren’t just standing on the pedals. With a system like the RM one the rider input and motor output torque are almost simultaneous. Feels really nice to ride, and wouldn’t be too hard to make as a DIY I think.

1 Like

No worries. Maybe, with a bike caliper mounted as it is no, there is no room inside it for the chain to go through, and running without lubrication would destroy the chain in no time

That being said, a reverse disc, with the caliper mounted from the inside, with a barrier to avoid oil getting on the pad section of the disc of a dry lubricant could work, but it still a bit tricky to hold the disc in place

This is a really nice solution, I’ve seen the folks from ebikes.ca using it on some custom bikes, having a torque based assist would be awesome, I will think about it

The torque sensing part is easy, the problem with all these ideas is the price of machining the components, even that can be made cheaper by just going laser cut parts, but one or two custom axles is unavoidable

There are a few helical gear sets on Aliexpress that could work

For the next version I will do the motor directly geared to the pedals using a #25 chain, gearing will probably be around 10:1 using a 9T and 90T sprockets, chain is on mail, I want to measure it first to be sure, all sprockets will be laser cut stainless

I want the bike bike working so I can get back to riding far, since I don’t commute right now and barely have places to go, none to be exactly due to corona, all will be bolted to the frame, the battery will be house in the triangle with a 3D printed and stainless combo enclosure, with room to the FSESC and this time space to play around if I want to add more things to the bike

1 Like

Incoming

Hopefully will have some time to put it together next week

1 Like

Guess I never updated here and completely forgot

The above setup worked reasonably well

But it has its problems and it’s not reliable

The main issue is that the the crank shaft isn’t rigid enough, so your chain tension changes as you pedal, and there isn’t a sweet spot or it’s tight enough to not skip but still loose enough to not have a ton of drag and too much tensions which will break the chain and motor

So the solution is to make a spring loaded tensioner, didn’t start yet because I still haven’t found a good way to do it

But first step was to replace the motor pinion, the laser cut ones worked but again not good enough, specially since I needed to make some 3D printed spacers to lock it in place, and that cracks way to easy

Got this comercial pinion which had a 6 mm hole, had to drill it and surprisingly it came out perfectly centered with a jig and tons of double and triple checking before drilling


4 Likes

So, I have a plan, now let’s print the plan again because I screwed up the first time since my laziness prevented me from going upstairs to get my calipers to measure the tube, and instead I used a ruler to do it, which went totally wrong

1 Like