First size test for the bike wheel hub battery I'm designing - which I'm going to tongue in cheek refer to as the Tri-Force due to its three pieces - is a success. There's enough clearance between the spokes to get the pieces in and out.
The plan is to attach them with straps so I can maneuver them, then secure them at the hinges. Depending on the angles I may be able to skip the hinges.
[#]bikeTooter #3dprinting #diy
=> More informations about this toot | More toots from yantor3d@mas.to
I'm enjoying the challenges of designing for the "Tri-Force" bike wheel hub battery.
Since the lights will be remote controlled, I need to fit a receiver into the plug; I don't want in the battery because that may complicate charging.
So, I have to fit a 11x24x8mm chip, and the eight wires involved (two from the battery, two from the lights, four from the chip) all into as small a form factor as possible, all in the space between two spokes forming a wedge.
Fun!
=> More informations about this toot | More toots from yantor3d@mas.to
After five and a half hours, the first full scale test is ready. And there are notes.
To accommodate using battery clips instead of a modeled holder and a nickel strip tail, I had to increase the thickness to 36mm. Now it barely fits between the spokes.
But, I was able to get in on and off with two of the three hinge pins in, which is good.
When the clips arrive on Tuesday I'll iterate on the thickness - I may have to use 21700 cells instead of 26650 cells.
[#]bikeTooter #3dprinting #diy
=> View attached media | View attached media
=> More informations about this toot | More toots from yantor3d@mas.to
The mark II Tri-Force bike wheel hub battery fits perfectly. The extra 4mm of clearance makes it easy to install with two of the three hinge pins in; although it helps to have three hands.
This means I have to go with 21700 cells instead of 26550 cells, which is fine. I think i need to recheck the outer diameter of the hub - 32mm is feeling a little snug.
Fortunately I already have a hinged ring designed for the handlebar mount, so I can reuse that.
[#]biketooter #3dprinting #diy
=> More informations about this toot | More toots from yantor3d@mas.to
The really cool thing about the lights I'm going to power with the wheel hub battery is they're remote controlled via a switch on the handlebars, rather than having to reach in between the spokes to turn them on/off.
I am not an electrical engineer but I feel it's probably a bad idea to have electrical components dangling freely in your wheels.
So I've designed a piece to go around the wires for the LEDs to house the receiver and the many wires involved.
[#]biketooter #3dprinting #diy
=> View attached media | View attached media
=> More informations about this toot | More toots from yantor3d@mas.to
A challenge in #3dprinting a bike wheel hub battery and controller will be the transmitter. I opted for an A23 battery for its voltage and form factor. Fortunately you can buy tiny little battery clamps to secure it.
Unfortuately there's exactly zero clearance between the battery and the base plate when installed, so I'm unsure how I'm going to make this work. I can't print in place because they'll interfere with the tool head.
I'm thinking a very small clip?
[#]biketooter #diy #electronics
=> More informations about this toot | More toots from yantor3d@mas.to
It's print day today for the bike wheel hub battery. I went through most of a spool of PLA iterating on designs, and now I can print the "final" parts.
I'm a little disappointed that I ended up having to have multiple shell variations - I could get away with just two if I had metric drill bits.
Oh well. I didn't design this for mass production.
[#]bikeTooter #3dprinting #diy
=> View attached media | View attached media
=> More informations about this toot | More toots from yantor3d@mas.to
A weekend of printing later and all the parts for the bike wheel hub battery - aka the "Tri-Force" - are ready.
The battery clips have pins that fit into slots and are glued in place. Everything else will be held together by nuts and bolts - a grand total of:
Too many fasteners? Not for an #ebike part!
[#]biketooter #3dprinting #diy
=> More informations about this toot | More toots from yantor3d@mas.to
"Ryan, you don't have to design everything to be a two identical parts with rotational symmetry."
"BLASPHEMY!"
It's an assembly jig for the battery case - to hold the halves of the three shells in situ while I do the wiring.
Why two parts? Because one continuous piece wouldn't fit on my print bed.
[#]bikeTooter #3dprinting #diy
=> More informations about this toot | More toots from yantor3d@mas.to
Let there be light!
I'm a little concerned that I seem to have fried the BMS on my test battery during this test - hence the pack of AAs. It did fine for the bench test a few weeks ago...
Oh well. The transmitter ia wired, and so is one of the receivers. I'll wire the other one tomorrow and see if I can pair it with the same transmitter.
Then we'll wire the battery cases, and once I'm sure it all works, then we'll order the LED strips.
[#]biketooter #3dprinting #diy #electronics
=> More informations about this toot | More toots from yantor3d@mas.to
I swear I recorded this with horizontal video! I don't know why it uploaded it as vertical...
=> More informations about this toot | More toots from yantor3d@mas.to
I tallied up the cost of building the bike wheel batteries last night.
$70 - and an extra $30 in shipping.
Sure, I could probably get that down if I could get everything in one place, but the full parts list includes forty eight screws, twenty four coupling nuts, twelve battery clips, six battery cells, two battery management system circuits, and two 2.1mm jacks.
I'm sure I might be able to get all of that on Amazon, but there's no telling the quality of it...
=> More informations about this toot | More toots from yantor3d@mas.to
And the battery cells themselves are half of the cost of parts. So there'd be minimal savings from redesigning the part to require fewer - or even no - fasteners or battery clips.
Ultimately, if the design even works, I want to put it out into the world. The parts are easy to acquire. There's not much soldering in the wiring - I taught myself in a weekend.
It's just the cases that a person needs... so I guess I should research print on demand solutions.
=> More informations about this toot | More toots from yantor3d@mas.to
No prototype survives contact with production.
The solder joints must be imperfect, because the light is temperamental and will turn on and off as I move things around to close the shells. The wires are interfering with each other, physically at least. There's too much happening in the main shell.
No worries, I can iterate on the design. I need to give more room for the wires...
=> More informations about this toot | More toots from yantor3d@mas.to
I was able to take it apart and put it back together again. Turning the center wedge 180 degrees (so the battery is "on top" when unfolded) meant for cleaner wire routes.
Tomorrow we'll charge it up, and do a bench test - see how long it can power the 12V LEDs I salvaged from a broken bike tail light.
=> More informations about this toot | More toots from yantor3d@mas.to
That feeling when you have a working prototype.
Half of one, anyway. The LED strips are on order and will arrive later this week. Then I need to decide if I want to adhere the strips to the rims or print dozens of spoke clips to make the lights less permanent.
The weight is good - approximately one pound / half a kilogram. We'll see how it stands up in motion next weekend.
[#]biketooter #3dprinting #diy #electronics
=> More informations about this toot | More toots from yantor3d@mas.to
I tried to get an instant quote from a print-on-demand shop for this and it says about $15 to print, and errored on shipping. Let's say $10. So at least $40 to print a pair of these, plus around $60 in supplies (plus shipping), most of which can't be found at your local hardware store.
I sure am good at designing complicated things...
=> More informations about this toot | More toots from yantor3d@mas.to
The mark I prototype is a successful failure.
Everything worked, for a while. The battery showed a nominal 12V and could power a salvaged e-bike tail light, and later a pair of 5V LED strips wired in series.
But then something went wrong - either the BMS shorted and took a cell with it, or vice versa. I have some (hopefully) better BMS boards on order and will build the mark II when they arrive.
I've also ordered a few options for the lights and will test them head to head.
=> More informations about this toot | More toots from yantor3d@mas.to
The two key metrics are brightness and longevity, with the latter being more important. I'd rather have modest lights that only have to be charged once in a while than overwhelming lights that have to be charged after a few hours of use.
The current contenders are:
=> More informations about this toot | More toots from yantor3d@mas.to
Spending a lazy Saturday iterating on the mark II of the Bike Wheel Hub Battery case.
This iteration requires fewer fasteners and has a more coherent construction history in Fusion 360. It also addresses the issue of the wires pinching between the shells.
[#]biketooter #diy #electronics #3dprinting
=> More informations about this toot | More toots from yantor3d@mas.to
I wonder if anyone has done the work to quantify the bed adhesion of #3dprinting based on total contact area - for both manual removal (a pulling force along the Z+ axis) and programmatic removal (a pushing force along the Y-axis)
Well, I've got nothing better to do today...
=> More informations about this toot | More toots from yantor3d@mas.to
One interesting trade off of printing at an angle is it doesn't like counter sunk screw holes - the "top" of the chamfer is technically an overhang so the slicer will warn about it.
Which is fine - it just means using different fasteners, or drilling your own counter sinks.
=> More informations about this toot | More toots from yantor3d@mas.to
The results are .... interesting.
The model printed at an angle was much easier to remove - too easy, in fact. The supports are better adhered to the bed than they are to the model.
The model printed at an angle has a uniform outer surface. The lower edge has some defects that could be remedied with additional support fins. There may be some droop in the holes and slots but they are still within tolerances.
[#]3dprinting
=> View attached media | View attached media
=> More informations about this toot | More toots from yantor3d@mas.to
Get in, loser, we're doing science.
One of the four options for my bike wheel lights has arrived. At least one other one should arrive today and then we can start doing bench tests and best of all, make a spreadsheet!
=> More informations about this toot | More toots from yantor3d@mas.to
Any good experiment requires a control. The baseline is the Wheel Brightz I currently have - twenty micro LEDs powered by three AAs in a narrow plastic tube.
I measure the lights with a light meter at the center of the wheel, and measure their draw using a multimeter.
Current: 5V / 4.7 mAh
Brightness: ~5 Lux
Vibe: Ok at night, in motion.
Link - https://brightz.com/products/wheel-brightz
Price - $15 / each
[#]bikeTooter
=> More informations about this toot | More toots from yantor3d@mas.to
First challenger is a two meter iNextStation "neon" LED tube.
Size: 12mm x 5mm x 2000mm
Power: 12V / 17 mAh
Brightness: 250 Lux
Vibe: Tron!
Link - https://a.co/d/4819hB9
Price - $10 / each
I'll probably need to 3D print some clips to attach it to the spokes with the optimum orientation. It has a 240° beam angle so it should be visible from both sides of the wheel.
If my math is right, I can get 200-250 hours of illumination on a charge.
[#]bikeTooter
=> More informations about this toot | More toots from yantor3d@mas.to
Typical... I gather several options for lights, and then discover another, potentially better option.
=> More informations about this toot | More toots from yantor3d@mas.to
Second lighting option in order of arrival and ease of testing is a length of LED rope light from Axel Lighting, a "marine and outdoor lighting" company.
This stuff is pricy, and a 27.5" wheel needs a little over seven feet. But it's meant for boats and Portland is wet, so I thought I'd give it a try.
Size - 3/8" x 86"
Power: 12V / 17 mAh
Brightness: 100 Lux
Vibe: Sparkles
Link: https://www.apexlighting.com/boat-lights/led-strips/led-ropelight-perfoot/
Price: $8 / ft
The lights are 1" apart, and will blur together at speed.
[#]bikeTooter
=> More informations about this toot | More toots from yantor3d@mas.to
Pros
Cons
I maybe have to modify the plug, but it's a good option. Visible from all angles, and the spacing means the there will be a dynamic element to the lighting when in motion.
It's out of the question to use if I go down the route of building and selling full kits - a 150' spool is $1,000 and would make 10 sets, so that adds $50 to the materials cost per wheel.
Yeah, no.
=> More informations about this toot | More toots from yantor3d@mas.to
I've got two more options to test - the EL wire is going to need some soldering so may not happen tonight. Next up is a pair of 5V COB LEDs wired in series. I expect it to be bright and thirsty.
=> More informations about this toot | More toots from yantor3d@mas.to
Surprise, surprise. The dual 5V COB LED strips are about the same as the 12V lights.
Size: 5mm x 2000mm
Volts: 8-9V
Power: 11 mAh
Brightness: 30 Lux
Vibe: Sleek
Link: https://www.superlightingled.com/super-slim-4mm-wide-5v-cob-led-strip-320ledsm-p-5707.html
Price: $15 / pair
Pros - lights on both side of the wheel, sharp, preciae appearance.
Cons - fiddly, may not stick well, slightly more complicated wiring.
[#]bikeTooter
=> More informations about this toot | More toots from yantor3d@mas.to
I'll solder the EL wire to the plug tomorrow - and I've ordered an option based on how well the "neon" LEDs act - a three sided diffuser that should give off more light.
And then I have to decide between the options, based on all the metrics I've listed.
I'm leaning towards the neon LEDs; so far they have been the brightest.
=> More informations about this toot | More toots from yantor3d@mas.to
The EL wire was a bust. I don't know if I damaged it while trying to solder it, or if the inverter is bad, but I couldn't get it to light.
Given how fragile the wires are, I don't think it would have survived being in bike wheels for very long.
=> More informations about this toot | More toots from yantor3d@mas.to
@yantor3d Note: some BMS cut off "permanently" when shorted out, but reset when charged. i.e. good luck resetting the unit in the field unless you have a second phone battery bank to trigger the charge mode with.
=> More informations about this toot | More toots from trouble@masto.ai
@trouble Aha! You are correct.
This is a 12V battery, so I don't know if a 5V phone battery bank would be enough to reset it...
=> More informations about this toot | More toots from yantor3d@mas.to
@trouble Although I guess if all I need is a brief 12V charge, I could build a reset stick with an A23 battery.
Edit: Oh, does it have to be fully recharged?
=> More informations about this toot | More toots from yantor3d@mas.to
@yantor3d nope, just needs to be woken up. My experience is with phone usb battery banks (wired two in series to get 10V; regulated. but don't mix your connectors up, lest you short out the pack!) However, USB-A is highly DIS-recommended. I had to replace several USB-A jacks because they couldn't take the vibration of being carried around on a bicycle.
=> More informations about this toot | More toots from trouble@masto.ai
@trouble Good to know. I've gone with 2.1mm for the jacks. And put them as close as I can to the center of rotation of the wheel.
=> More informations about this toot | More toots from yantor3d@mas.to
@yantor3d also recommend a dust cover. Rubber will work (quick and dirty) but get several, since you will drop and lose it. Moisture, dust, and rocks inside your connectors is bad. "DC Barrel rubber plug" will find the products.
=> More informations about this toot | More toots from trouble@masto.ai
@trouble I 3D printed the cases with a tongue and groove interlock around the perimeter, hopefully that will help keep them dry. I might do a bead of epoxy around the jack to seal it... I suppose I could print some sort of cover for the plug to give it some more protection when plugged in.
=> More informations about this toot | More toots from yantor3d@mas.to
@yantor3d I guess I'm recommending to not to overthink it. Rubber plugs are inexpensive. Try them, see if they get lost on the road due to vibration (depends which way they are facing). Then iterate. My bike hubs are always dirty, indicating the need for at least dust and road splash resistance. To test sealing experimentally, put glue on one half (left/right), and with your next iteration see if it leaked. Zip ties or string around the circumference might be a good safety cable just in case.
=> More informations about this toot | More toots from trouble@masto.ai
@trouble Good ideas! The cable will currently run through a case to house the receiver (remote control lights FTW!) with channels to clip it to the spokes. It should probably hold. If it doesn't I can print a new version with slots for zip ties...
=> More informations about this toot | More toots from yantor3d@mas.to
@trouble i wonder if part of my problem is I went with the only 12V 3S BMS I could find on Amazon, which means there's no way to verify the quality of them.
Searching a bit more I found exactly one option on Mouser... they look roughly the same but there's at least a brands behind them and instructions written in fluent English 🤔
=> More informations about this toot | More toots from yantor3d@mas.to
@yantor3d another option is a USB-c battery pack with a "trigger board" or module. Note that 12 VDC is optional in the USB power delivery spec, so you must search to find a battery pack that will output it. That said, you are planning for rotating batteries, for which a phone USB power bank is not optimal. So... Not actually helpful. For this project, anyways.
=> More informations about this toot | More toots from trouble@masto.ai
@yantor3d I can't tell, you're using stranded wire, right?
=> More informations about this toot | More toots from trouble@masto.ai
@trouble I think so - 24 AWG stranded. I'm tinning the wires before I solder them, but the battery clips don't seem to like to be soldered to...
=> More informations about this toot | More toots from yantor3d@mas.to
@yantor3d I would use larger wire. Automotive uses min size of 18 GA due to vibration and flex. 20 GA should be fine for this use case, except for battery shorting out (fuse can prevent melt down). Also consider batteries with solder terminals. Harder to get, but eliminates the springs as points of failure.
=> More informations about this toot | More toots from trouble@masto.ai
@trouble Noted. I'm looking at re-arranging the layout - inverting the middle shell - so the wires will all be on one side. I'll try a thicker gauge of wire next time.
=> More informations about this toot | More toots from yantor3d@mas.to
@yantor3d Is this to store the e-bike’s battery in the wheel, next to the hub motor?
From my days hanging out with racers, I remember them raving over light rims and tyres. Was that chase for lightweight wheels misguided, or does it not matter as much for an ebike’s rear wheel?
=> More informations about this toot | More toots from teotwaki@mastodon.online
@teotwaki Yes and no - it is to store a battery in the wheel, but not for a hub motor. I have a mid-drive motor with a downtube battery.
This is to power lights in the wheels themselves. I want to keep riding to work in the winter but we all know a driver's favorite song about cyclists - "Oh, I didn't see him!"
With three 21700 cells I can get a 12V 5Ah battery. I'm running some tests this week to figure out what kind of LEDs will give the best balance between brightness and longevity.
=> More informations about this toot | More toots from yantor3d@mas.to
@yantor3d That makes a lot more sense! Thanks for following up!
=> More informations about this toot | More toots from teotwaki@mastodon.online
@yantor3d Thanks for posting all this! It’s great to read about other people’s projects.
=> More informations about this toot | More toots from rasterweb@mastodon.social
@yantor3d Take a look at Slant3D on YouTube.
=> More informations about this toot | More toots from daryll@mastodon.social
@daryll Oh, I have watched basically every video on the channel. That's where I got the idea to print at an angle.
=> More informations about this toot | More toots from yantor3d@mas.to
@yantor3d I use a curved x-acto blade to get under one corner/end of the part then peel it off. The curve makes a huge difference vs straight blade to get underneath. Adhesion barely matters unless it's an over 3x3" flat bottomed part stuck to the bed. Yes, one should not pry with a blade. I mostly use it as a wedge, then use some other tool to finish the job. Small parts pop right off.
=> More informations about this toot | More toots from trouble@masto.ai
@yantor3d I built myself a metric screw collection (stainless, socket cap) from 2 to 8 mm over the past 2 years from kits on Amazon. The screws themselves are fine, but they are "thin", primarily the washers and nuts. The kits frequently have misleading contents like only one "cup" of screws is SS (closely inspect pics; they're accurate), or one of the cups is the wrong size (i.e. all 2.5mm except for this one cup of 2 mm). It's risky unless you're just generally building up a hardware store.
=> More informations about this toot | More toots from trouble@masto.ai This content has been proxied by September (3851b).Proxy Information
text/gemini
