Thats why i made that area thicker/stiffer, so flex wouldnt just be happening down by the hub, where there is the most leverage, but rather spread out along the length of the arm.The main problem isn't the thickness of the arms, it's stress risers which exist just about anywhere a sharp change in angle exists in a structure that flexes.
That's what breaks shafts and axles with circlip grooves, for instance. Even a tiny cyclic flexure induces stress at those points because taht's where the thing stops flexing, instead transferring the energy change (terminology?) to the material at that point.
If it doesn't break within the useful lifetime of the device it's part of, it doesn't. matter...
DIY (Mini) ebike/moped -> square tube frame, hoverboard motors - does this look reasonable? Any obvious oversights?
- Thread starter ziddan
- Start date
Your chain device is genius. And mesmerising to watch.
Your RS bearings should stay cool. I was looking at data sheets for different bearing types this morning, and the RPM ratings are surprisingly high for 2RS types. Your idlers will spin at probably only 2000RPM, depending on their diameter.
Reharding motor timing non-neutrality, I wouldn't get your hopes up. I haven't used them as left side drives, but I'm sure plenty of people do.
Your RS bearings should stay cool. I was looking at data sheets for different bearing types this morning, and the RPM ratings are surprisingly high for 2RS types. Your idlers will spin at probably only 2000RPM, depending on their diameter.
Reharding motor timing non-neutrality, I wouldn't get your hopes up. I haven't used them as left side drives, but I'm sure plenty of people do.
This conceptual diagram helps a bit to describe the challenges and benefits of incorporating changes that @amberwolf suggests.
The fillet cross section (right) distributes the stress more evenly, and lessens concentrated stress that can lead to part failure.
(Souce: Stress Concentration Factor)
The fillet cross section (right) distributes the stress more evenly, and lessens concentrated stress that can lead to part failure.
(Souce: Stress Concentration Factor)
Last edited:
Not really since the sprocket was very cheap and the fastener holes are drilled just a little off center
I do get that why doing fillets is a good idea, yes
Thanks, it was definitely the best quick idea i could come up with in the space constraint.
I DIDNT have my hopes set high for this, i just figure that trying it in reverse is being thorough.
BUT as soon as i did and sat on the bike, it takes off backwards with no issue at all, so i guess this motor is VERY directional.
Well, thats good as it is now working but annoying because i gotta redo the motor mount to turn it the other way round :/
ADDITION: The listing for the motor does specify it is to run counter clockwise but as ive seen multiple vids of people going over how to flip it be moving hall/phase wires, i expected they were doing it that way because their controllers didnt support reversing via the self learn wires and am now wondering if they have issues with cogging in that config or if messing with the wiring sorts the timing too.
My gut feeling is that it wont be the way to go because even if the sensor wires are shifter around, the "clock" positions of the sensors around the stator will remain wrong so i will just go with flipping the whole motor.
Last edited:
Ya know what, i think its horizontally symmetrical enough that i get away with not drilling new motor holes and the chain tensioner seems to have made the cutout i did for the chain on the right side unnecessary.
Gonna go have another test ride, hopefully with success this time.
Gonna go have another test ride, hopefully with success this time.
Did a couple of km around the vicinity and its a joy to ride.
Installed a speedometer app that would record top speed, since i didnt want to be looking at my phone for it.
On flat ground it goes to 26kph, almost exactly what my calculations had me expect.
Inclines are no issue, so that aspect is alright for going to and from work.
Cold fingers, big smile, Great success!
Now i need to figure out the range.
(And paint the frame, install the lights/reflectors and figure out some fenders so i dont get splattered if there is some wet on the ground.)
I guess ill design some mounts for the lights.
ADDITION: as it is currently:
Installed a speedometer app that would record top speed, since i didnt want to be looking at my phone for it.
On flat ground it goes to 26kph, almost exactly what my calculations had me expect.
Inclines are no issue, so that aspect is alright for going to and from work.
Cold fingers, big smile, Great success!
Now i need to figure out the range.
(And paint the frame, install the lights/reflectors and figure out some fenders so i dont get splattered if there is some wet on the ground.)
I guess ill design some mounts for the lights.
ADDITION: as it is currently:
Last edited:
Knocked together a very simple mount of the large LEDs i picked up for front/rear lights, this one is intended for the rear.
The LEDs have a threaded tube and a nut to hold it, the mount is a clamp design which will utilize that nut to hold it to the frame rail.
Its angled so it will be horizontal when attached to the rearmost tuve on the left side, gonna give it a print and see if i like it, if not ill do something else.
Not sure if this would work up front, im gonna look at it some more after work.
The LEDs have a threaded tube and a nut to hold it, the mount is a clamp design which will utilize that nut to hold it to the frame rail.
Its angled so it will be horizontal when attached to the rearmost tuve on the left side, gonna give it a print and see if i like it, if not ill do something else.
Not sure if this would work up front, im gonna look at it some more after work.
How are you going to get it over the frame tube?
Im hoping its going to be flexible enough to just force it apart ove the tube.
The section through which the LED tube goes has a cut all the way throught which will be held closed, in clamping action, by the nut.
There is very little infill in this part and none where it needs to flex, as it is mostly "wall" in 3d slicer terms.
As for flexible infill, that feels a lot like hype to me, except i very special cases.
Most of print strength is in tops/bottoms/walls of a print, most of what infill does is to be something to put the top on.
It did flex enough
Just got in from another test ride.
When i got the batteries they were reporting full charge via the build in LED indicator, after the drive yesterday i put one of them on charge, ive only got one charger at the moment, switched it to charge the other battery and decided to see what would happen on just one battery in the bike.
In order not to overdraw it i put the controller current limit down to 9 amps, which should be about 0.8c.
There really isnt a noticable difference in how it drives, i guess the gearing makes regular driving a breeze for the motor.
On climbs id see the battery stack on the display go down but once back on flat ground it would rise again after getting up to speed.
Decided to do a torture test by driving up the steepest longest hill in the area, colloquially called "Mördarbacken" / "The killer hill" by locals.
I recall having to downshift to first gear when driving up it on my Zundapp KS50 back in my teens, but it was fine on this DIY, sure it wasnt the fastest of climbs but it got up without issue.
Gonna have a start at designing a front light fixture while hoping we get some warmer weather soon, its current 4C here so my hands are cold again after the ride
When i got the batteries they were reporting full charge via the build in LED indicator, after the drive yesterday i put one of them on charge, ive only got one charger at the moment, switched it to charge the other battery and decided to see what would happen on just one battery in the bike.
In order not to overdraw it i put the controller current limit down to 9 amps, which should be about 0.8c.
There really isnt a noticable difference in how it drives, i guess the gearing makes regular driving a breeze for the motor.
On climbs id see the battery stack on the display go down but once back on flat ground it would rise again after getting up to speed.
Decided to do a torture test by driving up the steepest longest hill in the area, colloquially called "Mördarbacken" / "The killer hill" by locals.
I recall having to downshift to first gear when driving up it on my Zundapp KS50 back in my teens, but it was fine on this DIY, sure it wasnt the fastest of climbs but it got up without issue.
Gonna have a start at designing a front light fixture while hoping we get some warmer weather soon, its current 4C here so my hands are cold again after the ride
Last edited:
I'll trade you for our current weather (gets hotter from this point): (is in F not C, but the high expected friday would be nearly 39C)Gonna have a start at designing a front light fixture while hoping we get some warmer weather soon, its current 4C here so my hands are cold again after the ride
I havnt and for now i dont intend to, speed is just over the intended goal and its really torqey as it got up that steep hill, something id like to keep
That will also be a no, im happiest in the 15-25c range but do prefer it below, rather than above, that range if outside of it
Yes, but I defer to my first post in your thread. Because you have a beefy motor you can regain the torque via higher amps (when you get a more capable controller and battery) and have quieter and longer lasting drivetrain. 5kg of motor provides flexibility, you don't need high RPM/reduction. As you've already discovered, there's torque there, and plenty more you're leaving on the table ... you're feeding the motor half the current it can comfortably handle.
It's not so much the noise from the 9T sprocket, like I said earlier I somewhat like it (with the chain travelling 3X faster than yours), but the wear rate.
Still, great to hear it's meeting expectations.
It's not so much the noise from the 9T sprocket, like I said earlier I somewhat like it (with the chain travelling 3X faster than yours), but the wear rate.
Still, great to hear it's meeting expectations.
The design for the headlight is done and on the printer, its a simple clamp around the headtube, exit on the back for wires.
That climb did get quite slow and i would doubt it could manage if geared to be quicker, rather than this strong.
As for the weekend plans, supposedly its going to be around 10c without rain tomorrow so im aiming to run one of the batteries down to test for distance, while keeping the other one in reserve to get back.
And tips for free android apps that will track gps speed and distance travelled?
And i want to keep it
That climb did get quite slow and i would doubt it could manage if geared to be quicker, rather than this strong.
As for the weekend plans, supposedly its going to be around 10c without rain tomorrow so im aiming to run one of the batteries down to test for distance, while keeping the other one in reserve to get back.
And tips for free android apps that will track gps speed and distance travelled?
OsmAnd.
Thanks, will check it out!
I decided to redo the rear light(s) by combining them into one, instead of having a LED on each of the rear tubes, more in line with the style of the headlight.
Knocked together a quick design for a phone mount to sit in the frame square in front of the saddle for the test ride and possibly longer.
Added some screw holes in the blank space so i might easily add on a panel for whatever, possibly a light switch initially but i think im likely to do something more firmly attached for that.
Added some screw holes in the blank space so i might easily add on a panel for whatever, possibly a light switch initially but i think im likely to do something more firmly attached for that.
Okay, the weather today isnt all that much warmer but long johns and gloves do wonders to keep the chill at bay.
Did as planned and ran on one battery.
After 8km i started to notice a drop in power and rode on for another 4km, until power was waaay down but giving a total of 12km on one of the batteries.
I was getting worried that the LVC wasnt working so i switched to the other battery and went home easily enough as i had been doing a circuit, not going straight in one direction.
When i got back i multimetered the down battery and it reads 35.6v, which is quite a bit above the LVC.
So with both batteries combined, sharing the draw as the voltage gets lower, i think i could reliably go 25km.
Easily enough to go the 16ish km to work but most likely requiring a charge to get back home.
I might also add another lead to the parallel adapter so i can use the 4.4ah hoverboard battery along with the two ebike batteries for some additional range.
Ill have to make a splitter for my charger so both batteries can get trickled into as i may be off on assignments and unable to switch the charger between them at work, maybe get another charger and make two splitters so i can have one at work and not have to bring the one from home.
Ive decided against a splitter and ordered another charger.
Average speed was 17kph and top was 33kph, 110m uphill and downhill.
All in all a successful test IMO.
Did as planned and ran on one battery.
After 8km i started to notice a drop in power and rode on for another 4km, until power was waaay down but giving a total of 12km on one of the batteries.
I was getting worried that the LVC wasnt working so i switched to the other battery and went home easily enough as i had been doing a circuit, not going straight in one direction.
When i got back i multimetered the down battery and it reads 35.6v, which is quite a bit above the LVC.
So with both batteries combined, sharing the draw as the voltage gets lower, i think i could reliably go 25km.
Easily enough to go the 16ish km to work but most likely requiring a charge to get back home.
I might also add another lead to the parallel adapter so i can use the 4.4ah hoverboard battery along with the two ebike batteries for some additional range.
Ive decided against a splitter and ordered another charger.
Average speed was 17kph and top was 33kph, 110m uphill and downhill.
All in all a successful test IMO.
Last edited:
Having a go at making a front fender that will attach to the v-brake posts.
The main part of the fender is split in two down its length for ease of printing and replaceability if a side gets damaged, joined in the middle using an overlapping part and then screwed into the v-post mount.
The main part of the fender is split in two down its length for ease of printing and replaceability if a side gets damaged, joined in the middle using an overlapping part and then screwed into the v-post mount.
Dont have any black filament, which i wanted for the mount and the store was out when i visited on friday, so it will get sprayed eventually.
Need to get some correct length screws but otherwise im pretty happy with it.
Going to whip up a rear fender based on this one, its going to be slightly longer with a small upsweep at the back and i might split it differently.
Need to get some correct length screws but otherwise im pretty happy with it.
Going to whip up a rear fender based on this one, its going to be slightly longer with a small upsweep at the back and i might split it differently.
Last edited:
