|Campus Rider MK I.5| Weld-free Electric Long Board

[spinningmagnets]

Awesome build! On Keplers super-commuter, he machined a housing out of high-density-poly-ethylene (HDPE), and it cut like butter, very easy. I wasn't up to date on the best candidates for a power board, so I immediately noticed the D-shaped axle on this model of truck...I would definitely spec these if I was making a similar project.

I think my board maxes out around a 10-15% grade. But I only have 6S, so I don't know what a better power system would be able to do

If the drive wheel is starting to slip when you are on the steepest incline, you need 2WD, so maybe try a second system on the other axle. If it has enough traction, but it's starting to bog down then you need more power, I'd go to more voltage. I wouldn't go faster than 20-MPH, so if you go to 10S / 37V, spec a motor with a 30% lower Kv, somewhere around 180-Kv. If you double the volts to 44V, then you can cut the Kv in half (I know 63mm motors can be found in a 149-Kv, I wouldn't go smaller than 63mm diameter).

For more speed without changing the volts or Kv, I'd make the drive pulley larger to keep as many belt-teeth engaged as physically possible (rather than making the wheel-pulley smaller).

Well done. The water-jetting can be much cheaper per part if you order 5-10, so consider making up several kits and selling just the ones you don't need. This would also let you try 2WD, then sell it then you'd be back at one-wheel-drive. You could easily sell 4 kits to pay for the 5th.

 

[lpbug]

What a guy! Awesome man! Should be beneficial to the people who have access to a machine shop or waterjet.

Lpbug - What is the size of the bolt that you use through the drive wheel cog?

Was planning on using m3 but couldn't find that at my local hardware store... Had to resort to using 4-40. Although it's smaller, it seems to be holding up well. The next prototype will feature m4 rods for added rigidity.

 

[spinningmagnets]

Another quick note, you should try to find someone who has a Castle Creations ESC that might let you borrow it. They have data-logging, and a quick spin up your steepest hill will log the max amps drawn. I suspect you would be fine with an 80A ESC, but regardless of the amp-capacity, I also recommend adding two 50V low-ESR capacitors across the power leads (I'll add a link tonight). They are only $2 each, and they can save your ESC from frying under some conditions. Cheap insurance that is highly recommended.

 

[lpbug]

Hi Magnets- thanks for the input on the HDPE enclosure. I will see if that's a viable option.

Another update: I will work on MK I.5 after finals are over. Upgrades will include the following...

-(Obviously) Working enclosure to add much needed weather resistance
- On-board voltage display
- Modular compatibility with smaller boards (Pocket Rocket)
- Arduino "pseudo-compatability", meaning I consider the integration of an Arduino in my MK II "Smart rider" build. Work out details for integration and interfacing of an actual display, current+voltage sensor & accelerometer. It would be nice to keep these sensor modular as well since they are pretty small. Perhaps just a small "black box" like thing I can hook on.

What else do you guys think is missing?

 

[spinningmagnets]

Here's the link to the low-ESR 50V capacitors. The two legs are marked pos/neg, and you just solder them onto the two wires going into the ESC. They absorb and smooth out any voltage ripple.

http://endless-sphere.com/forums/viewtopic.php?f=30&t=22194&start=25#p382837

Here's keplers HDPE machined plastic housing:
http://endless-sphere.com/forums/viewtopic.php?f=6&t=47139&start=50#p698485

 

[RiffSurfer]

What a guy! Awesome man! Should be beneficial to the people who have access to a machine shop or waterjet.

Lpbug - What is the size of the bolt that you use through the drive wheel cog?

Was planning on using m3 but couldn't find that at my local hardware store... Had to resort to using 4-40. Although it's smaller, it seems to be holding up well. The next prototype will feature m4 rods for added rigidity.

I'd recommend 10-32 bolts

 

[lpbug]

What a guy! Awesome man! Should be beneficial to the people who have access to a machine shop or waterjet.

Lpbug - What is the size of the bolt that you use through the drive wheel cog?

Was planning on using m3 but couldn't find that at my local hardware store... Had to resort to using 4-40. Although it's smaller, it seems to be holding up well. The next prototype will feature m4 rods for added rigidity.

I'd recommend 10-32 bolts

I am actually having very little issue with 4-40. Since rigidity is exponentially related to thickness, I think bumping up to M4 threaded rods will be sufficient. Additionally, the hub stops need enough clearance and I think m4 is the maximum bolt size I can put in there without compromising its structural integrity. If I end up needing more, I could just add more hub stops+rods. The beauty in a modular build eh?

 

[RiffSurfer]

I am actually having very little issue with 4-40. Since rigidity is exponentially related to thickness, I think bumping up to M4 threaded rods will be sufficient. Additionally, the hub stops need enough clearance and I think m4 is the maximum bolt size I can put in there without compromising its structural integrity. If I end up needing more, I could just add more hub stops+rods. The beauty in a modular build eh?

Very true. Where I work 10-32 is the most common bolt size, and it fits in there almost perfect. The nuts however are a bit tight, and I had to file down the squared edges of the calibers a bit.

 

[Murfix]

Hi Magnets- thanks for the input on the HDPE enclosure. I will see if that's a viable option.

Another update: I will work on MK I.5 after finals are over. Upgrades will include the following...

-(Obviously) Working enclosure to add much needed weather resistance
- On-board voltage display
- Modular compatibility with smaller boards (Pocket Rocket)
- Arduino "pseudo-compatability", meaning I consider the integration of an Arduino in my MK II "Smart rider" build. Work out details for integration and interfacing of an actual display, current+voltage sensor & accelerometer. It would be nice to keep these sensor modular as well since they are pretty small. Perhaps just a small "black box" like thing I can hook on.

What else do you guys think is missing?

Hey man, I would like to upgrade my board with a display as well. Alas, I'm no electronics expert. I don't know about your skills, but I'd like to figure this out together with you (and others). I'd like to have more than just voltage display though. Lemme know!

Good luck with them finals!

 

[michaelkramer11]

Anyone planning on getting any of these cut and milled?

I'd be interested in a set if anyone is.

 

[arjanmobin]

alright. Thanks for posting the files. Also I had another question. The hobbyking website said that the motor draws 65 amps. Why do we need an ESC that is 150 amps? Isnt that REALLY overkill?

 

[Murfix]

alright. Thanks for posting the files. Also I had another question. The hobbyking website said that the motor draws 65 amps. Why do we need an ESC that is 150 amps? Isnt that REALLY overkill?

It's always best to have a safety factor. Also, motors can draw very high currents when bursting: up to twice or more the max continuous current.

 

[lpbug]

@ Murfix- Cool, I have worked with plenty of Arduino projects. Most of my experience comes from working with quadrotors. I built these from scratch so had to work with/learn CAD/CAM, some embedded programming, and standard workshop skills. It shouldn't be too hard to integrate a display like this one or even a smaller one. I have to work out where best to mount it though. One viability is to profile a small hole on the board and mount it underneath (but this is risky when the board undergoes substantial stress as the flex might snap the lcd, unless it is not rigidly mounted). Another is just to have it on the side of the enclosure, but it won't be as easy to glance at the display which can display useful information such as voltage, mAh consumed, current milage, speed, etc... I'll still have to work out the details.

@arjan- Along with what Murfix said...As far as I know, the current rating they give you is measured at just a little under maximum voltage. Use the power equation instead: (Voltage*Current=Power). Look at the motor's maximum power then divide that by the voltage of the pack you're running, and you'll be able to get the max current you can run through the motor to not exceed the power cap. So if the motor's max power rating is 2000W, I can take my 6s pack (~22.0v let's say) and find that Currentmax=2000W/22v~91A. While you'll probably never draw that much continuous power, nothing is to say you won't do that bursting (for short durations). If you are really that concerned, you should get the overrated ESC first, then through testing find the max current that you draw, then buy another ESC that is more closely matched.

 

[Murfix]

Awesome dude!

I think the only place to put a display is right in front of you. On top of or below the deck like u said. How are you going to measure all the variables like voltage, current, ...? I guess programming the logic isn't the hard part, for me it would be setting up the circuit and measurement tools.
If I can do anything to help, let me know! I'd love to get into this.

 

[lpbug]

Awesome dude!

I think the only place to put a display is right in front of you. On top of or below the deck like u said. How are you going to measure all the variables like voltage, current, ...? I guess programming the logic isn't the hard part, for me it would be setting up the circuit and measurement tools.
If I can do anything to help, let me know! I'd love to get into this.

The placement of the display is still something to decide on but I can measure all the variables by integrating sensors. For voltage, a simple voltage divider should do the trick since I'm not asking for +-0.00001V here. The current can be detected with a current sensor, I will probably get one that is capable of handling at least 150A, just like my ESC. The speed of the skateboard can be deduced from the RPM of the wheel via an photosensor assuming that you are not slipping along the surface (pretty safe assumption, most of the time at least). Additional sensors include GPS receiver and accelerometer, but those will likely be integrated later.

 

[spinningmagnets]

The hobbyking website said that the motor draws 65 amps. Why do we need an ESC that is 150 amps? Isn't that REALLY overkill?

''

Model cars and airplanes accelerate very rapidly, so the motor-amps don't peak as high or for as long as they would if they were pulling a 160-lb / 73-kilo person. Nothing in the catalog was written in reference to E-bikes, or E-boards...This is also the reason you need additional capacitors, where a model car/plane would not.

 

[dirkdiggler]

My RC cars hit way higher amps than any of my rideable projects. When you drive an RC car, you go full throttle to stop over and over. This is extremely hard on the ESC. The amp spikes on a board are fairly level. You can't accelerate quite as fast as the RC car does. The 150a car ESC is made to take that 20 lb car to 60 or so MPH in seconds. No way you want to do that on an eboard. The burst current is rated at 1080amps for the HobbyKing version. The spikes I get from my eboard are on hill climbing, but never really get over 100a on 6s. Thats over 2400 watts. I know ebike controllers have more fets, but they are setup for much lower amps. 1000 watts on an ebike is more than most need. At 48v thats only 20 amps. That is one of the reasons an HV ESC would be nice to have, but the expense and hassle isn't worth it. I've ridden the 90a boat ESC with no issues all summer. The burst current on that is only 120a. I still think the ESC's we are using are more than needed. The main reason the 150a is used is because of HumboldtRC. He blew his 120a, but he was doing 55mph on a mountainboard and riding like a wildman. The other issue is you need an ESC for 6s. His 120a esc is made for 4s. I can't recall if he was using 4s or 6s.
You are correct that the motors are only rated at a max of 60-70 amps anyway.

 

[Murfix]

Awesome dude!

I think the only place to put a display is right in front of you. On top of or below the deck like u said. How are you going to measure all the variables like voltage, current, ...? I guess programming the logic isn't the hard part, for me it would be setting up the circuit and measurement tools.
If I can do anything to help, let me know! I'd love to get into this.

The placement of the display is still something to decide on but I can measure all the variables by integrating sensors. For voltage, a simple voltage divider should do the trick since I'm not asking for +-0.00001V here. The current can be detected with a current sensor, I will probably get one that is capable of handling at least 150A, just like my ESC. The speed of the skateboard can be deduced from the RPM of the wheel via an photosensor assuming that you are not slipping along the surface (pretty safe assumption, most of the time at least). Additional sensors include GPS receiver and accelerometer, but those will likely be integrated later.

Cool dude!

You could easily tape something reflecting on the wheel sprocket and mount a photo sensor on the hanger. Keep me posted

 

[Texonite]

Why not use simple bicycle-computer magnetic sensor?
Optic sensor will get dirty at some point...

 

[kkEdlund]

Has anyone tried to download and open the CAD files? Maybe I'm just being retarded (it certainly wouldn't be the first time), but as far as I can tell Solidworks eDrawings believes that the truck mount is in excess of 60 inches tall. I can either rescale it or try to find a bigger waterjet (lol), but has anyone else had a similar issue?

EDIT: If I am eye-balling this correctly, it looks like it is about 25X bigger than it needs to be. lpbug, any chance you drew the part in mm instead of in?

 

[lpbug]

Has anyone tried to download and open the CAD files? Maybe I'm just being retarded (it certainly wouldn't be the first time), but as far as I can tell Solidworks eDrawings believes that the truck mount is in excess of 60 inches tall. I can either rescale it or try to find a bigger waterjet (lol), but has anyone else had a similar issue?

This was CADed in metric- perhaps try playing with scaling down by a factor of 24.5? Also, I have MK I.5 CAD files coming up after finals (around Christmas), which features an improved design for the truck mount if you would like to wait around before you cut it.

 

[Murfix]

I know Creo 2.0 has the option to interpret units. 25 will still be 25 but then inches instead of mm so you don't need to rescale. Dunno about Solid

 

[kkEdlund]

This was CADed in metric-

Yup, that'll do it. Thanks. I have solidworks set up to use IPS (Inch, Pound, Second) units, so when eDrawings sees a .dxf and it isn't sure what the units are, it applies the defaults. Re-scaling the part, or changing my default units should correct the problem.

I'm working on a one-piece mount, and was just curious how some of your geometry looked. Nice work, by the way.

 

[lpbug]

A quick update: It's been snowy here in Boston but I've been able to catch a ride on the board a few times here and there. Nothing as intense as it was in Fall but the board is holding up great. I still haven't used any locktite on my motor pulley (also only using one of the two set screws) and it hasn't slipped, which leads me to believe that the gear ratio is a bit too conservative. I feel like I could get even more speed without frying my motor (but it will definitely be risky using the XT-60s.

I have a new idea for the enclosure and will try to use the waterjet for it soon. I am taking multivariable calculus as well as doing research for autonomous drones on the side so time is kind of scarce. Once I get the finer details of the board sorted out and integrate everything I might set up an online store for kits that I can waterjet for those that do not have immediate shop access. Pricing should be reasonable and below most other commercially available options.

 

[lpbug]

It has been quite a while since I've posted any updates regarding this build. Time is just not something you have in college especially with the ridiculous engineering classes and research. I had (and still have) many plans for improvement on this board. However, there have been issues that required more attention than I paid them. First of all- Enclosure. Having a lipo exposed to the elements is not a very smart idea. Riding a board with an exposed lipo through rain and salted roads is even worse. I did both, but hey, gotta live on the edge right? But seriously, enclose the electronics as soon as possible, I probably wore out my components by half their lifetime by not doing this. I finally got a weekend free and went to the lab I'm working in and waterjetted out some pieces of 1/4'' aluminum to make a simple box. It came out pretty well and I used double sided tape on the walls to stick it on the board.

It will be screwed on eventually, but for now double sided tape does fine. Important thing to note here: I implemented ventilation holes so the enclosure is not completely insulated. The ventilation hole diameter is exactly 1mm. Tiny enough to be splash resistant, but there are enough of them to facilitate actual air flow.

Also, I did some work with the ABEC 11 wheel drive hubs, which lead to an unexpected discovery. So there's nothing wrong with my original hub stops, but they could be tapped to allow the use of bolts instead of threaded rods. In my opinion, the bolt option is less hassle. By using hub stops, I retain the modular design aspect as well as simplify the machining process. The main challenge in this step is getting the dimensions of the geometry for the hubs. I tried contacting several of the members here who seemed to have pretty nice fitting ABEC 11 hub drives but it's either no response or incompatible file. Since I'm still in the process of building my own 3D printer, I decided to take a more classic approach. I also ended up with something that looks like an ancient crumbling Greek temple or something of that nature.

This model was obtained by pouring plaster in one of the ABEC wheels I had. The part turned out great except I broke all the prongs off while removing it. No big deal though, I had complete pieces of the hubs and could derive everything I need. Using calipers, I recreated the geometry in Autodesk Inventor and processed that to be waterjetted (files to be posted later).

I ended up with these, which I am pretty pleased with. The fit is absolutely spot on and there is no wiggle room in the θ̂ direction.

I tapped three of these and am now using M3 screws to keep the drive in place. This simplifies from my last build a bit (I think) because working with threaded rods just kind of got a little tedious. Anyway, I was very surprised when I realized that the hub geometry of the new wheels that I got (ABEC 11 clones) are different from the previous set of clones I had. Same wheel diameter, same eBay supplier, but clearly different hub geometry. I might just buy a set of real ABEC 11s to see if the hub geometry for that is different too!

This has been quite a long update, but I am likely not going to get time to work on the board for the next few weeks (exams, research, 3D printer, life, etc). Future updates include a magnetic snap-on top as well as integration of an Arduino to interface a wireless Wii nunchuck with the ESC and later to house other sensors such as voltage meter, GPS, etc... I have an interesting safety feature that I would like to implement with an accelerometer sensor in the future.

Note: A few of you have come forward and notified me that the files are not there anymore. Rest assured, I am going to find a more reliable site for hosting the CAD files and post the MK I.5 files as soon as possible after I consolidate them.