The purpose of this project is to learn and have fun. I love designing things, and this project provided a fun and challenging way to make something useful.
My goal for this product was to create an easy to use, high powered, long ranged, good looking electric longboard
I started with a set of criteria that I knew I wanted for the board (left). Using these criteria, I researched and developed my own solutions to accommodate each requirement (right). In addition to my criteria, natural requirements arose (bottom left).
Balanced power in each direction
Fits my current longboard
Easy to use
Components do not touch ground
Batteries supply sufficient current and voltage for the motors
Power must be delivered efficiently to wheels
Two motors, one for each rear wheel
Add batteries in parallel
Motors fit behind the rear wheels, batteries fit under the board
Minimize number of actions to switch from use to charge
Solutions for Natural Criteria
Find thin batteries; 3D printed casings for components
Using motor specs, determine required battery combination
Use timing pulleys to offset the motors from the axle
After some preliminary research to determine the overall look and shape of my longboard I drew up a quick concept drawing.
This drawing features 6 batteries (6s LiPo's) encased in partially visible 3D printed plastic housings. The motors are tucked under the board and power the two rear wheels over a timing belt system. The ESC's sandwich the base of the T shaped battery configuration at the rear of the board. Headlights and tail lights with the addition of a concept logo showcasing a "jkboards" logo complete the look.
I started my search with the motors. Once I found the motors, I could determine the specs for the rest of the components. Based on the motors, I determined that I would need a circuit that supplied 44 volts (2x 6 cell LiPoly's in series). I found applicable batteries, and I wanted more range, so I designed a circuit to house 4 batteries: 2 in parallel in series with 2 in parallel. This complicated the battery charging situation because these batteries must be charged either one at a time or in parallel. Because my goal for this project is ease of use, I designed a schematic that allows me to switch the circuit from powering the motors in semi-parallel to charging the batteries in full parallel with relatively few actions for each switch. Shown is a sketch of this circuit.
After I had the circuit drawn, I pulled out my computer and drew up scale diagrams of the board for sizing and visualization of the board. Shown above is the 2nd rev of circuit diagrams with the same schematic as shown in the Post-It above. Because it is hard to find switches that operate at the high amperages that I am working with, I will be using what are known as loop keys, which are basically two wire plugs turned into a circuit closing device. The scale diagrams are seen at the top of this page next to "purpose" and "goal." Below is a comparison of required actions without a complex circuit versus with this circuit.
Unplug battery 1
Remove battery 1 from housing
Unplug battery 2
Remove battery 2 from housing
Unplug battery 3
Remove battery 3 from housing
Unplug battery 4
Remove battery 4 from housing
Plug in Battery 1 (to charger)
Plug in Battery 2 (to charger)
Plug in Battery 3 (to charger)
Plug in Battery 4 (to charger)
Unplug loop key module from port 1
Insert loop key module into port 2
Insert charging cable
After I had determined what parts to get, I started working on a CAD of the Longboard to help design and visualize what the final product will look like. I started by removing the front hanger from my longboard and taking precise measurements of specific points to create a 3D model of the part. I did this for the three main parts that make up double king-pin trucks, as well as the major parts that make up a skateboard, like the deck and wheels. From there, I designed a mounting mechanism for the motors, then discovered the motors will not fit under my board with enough clearance of the ground. I switched their orientation and designed a three-plate mounting system for the two motors to keep them at the desired distance for the timing belts and pulley's. Based on existing pulley CAD's I designed custom pulley's that can be machined on a CNC mill. To attach my wheels to the driven pulley, I plan to create holes in the wheels with a CNC mill to line up with my custom pulley. Renders of my progress are shown at the top of this page.
The next steps are as follows:
Review motor and ESC power ratings
purchase motors and ESC's
Test and prototype motor mount design
Purchase wires and (XT90) connectors
Solder according to schematic
Design and build/3D print housing for batteries and ESC's
Test using old RC radio system
Purchase or design handheld remote
Design mounts for powerful headlights, side reflectors, and brake/taillights
(one year later)
After saving up enough money to buy enough components to build a 'minimum viable product' (e.g. one that would get me started, and that I can ride around campus), I went ahead and purchased the following components:
2x 6355 260kv motors
2x 50A TorqueBoards speed controllers
1x 'nano' receiver and remote
2x motor mounts for TorqueBoards trucks
4x 83mm ABEC 11 Wheels
2x aluminum motor pulleys
2x aluminum wheel pulleys
2x timing belts
4x 6s Zippy Compact 5000mAh batteries
High Strength Velcro
10 AWG silicone wire
Getting it Moving
I was excited to get the board moving, so I threw together all the components I bought, which included plenty of soldering, and some filing to get the mounts (which weren't designed for these trucks) to fit on my dual kingpin Gullwing Sidewinder II trucks. These trucks are fantastic trucks that are stable at higher speeds around 25 mph, yet still allow plenty of carving, which is absolutely necessary when riding around a college campus. I clamped down on the trucks with the set screws on the motors mounts, and realized that there wasn't enough room on my 9" sidewinders to fit both motors. I tried it out with one motor attached and realized that the whole system couldn't even push me. Something was wrong, but I didn't know what.
I ordered the 10" sidewinders, which are the same dual kingpin design, just wider. I immediately used a file to create flat notches for the monstrous set screws that hold the motor mounts to the trucks, then attached both motors. There is not enough room under my board for the motors to fit toward the center of the board, so I mounted them facing the back of the board, as you can see in the image above.
The way that the bolt pattern worked on the motor mounts didn't allow for me to use the drop through in the deck. I used top mount to get things moving.
Instead of using all four batteries, I decided when the components all arrived to just go with the simpler route, and use only 2 batteries in parallel. This was effectively a 6s system, running at 22.2V nominal with 10,000 mAh (and way too many amps available)
I didn't have a proper enclosure for the batteries and electrical components, so I used an old cardboard iPad box, which was the perfect size. I used bungee cords to hold it to my board. Everything, including the batteries, was mounted toward the rear of the board. This made the center of balance quite uneven.
I used my mac running VESC tool on windows via bootcamp to reprogram the VESC. Since the speed controllers shipped with 12s as the default expected voltage, the board thought it should be in low voltage cutoff mode, which provided barely any power at all. I changed the setting to 22.2V nominal, or 6s, and the board finally started, and it packed quite a punch too. My friends who tried it were knocked off their feet. Literally. The punch was a little too powerful, so I adjusted the remote/response curve settings. This made throttle control much smoother without sacrificing the power that the board could deliver.
After working up my confidence, I probably got to about 15 mph. There really was no way for me to tell at that time, and I didn't need to know. I just knew that the board was fast enough for me to get around campus, though I was testing in the parking lot of the elementary school in my neighborhood.
It worked, but it still had plenty of room for improvement. To spare you the time of reading through all of the upgrades and improvements and decisions (and me the time of writing and recalling all of that), I will just describe the 'final' product. That is, how the board ended up a year later.
P.S. Check out the latest video here of my sister trying out my board for the first time!
The black wheels are "80a" but they don't feel like it. They give a super rough ride. Not comfortable at all. I did some research and ended up buying the same wheels that boosted used. I also bought some 32T pulleys from evolve to fit those wheels. They are the 80mm Orangatang Kegel's. They are much smoother than the 83mm wheels I bought with the kit of parts I purchased earlier, though they are supposedly the same durometer. After a year of cruising, I want to upgrade to the 85mm Caguamas by the same maker, Orangatang, though I do like the lightweight low profile 80mm, since they are perfect for around campus, but they don't make for a great street cruiser.
I realized the batteries I had had plenty of range. Too much, in fact. I could go a whole week on campus without charging my board. I decided to sacrifice a bit of range in favor of a new battery system. I decided to stick with a 6s system, as, at that time, it still gave me plenty of speed. I decided, however that instead of using one 6s battery to provide the 22 volts, I would use 3x 2s batteries. I used 3 6600 mAh Turnigy Nano-Tech hard-case batteries due to their extremely low profile, which meant plenty of ground clearance. When mounted to the underside of my board, I could still go over even the most aggressive speed bumps without a scratch. I used flat speaker wire to seamlessly connect the battery on the front of the board to the motors and ESC's on the back. I connected an anti-spark power switch to the ESC side of those wires.
This meant that those wires were always live (bad idea). When soldering another component, my line of solder contacted both wires simultaneously, and caused a very bright spark and loud bang. Everything, however, seemed ok afterwards because the shear power from the batteries just completely blew the solder wire into the air, and disengaged the accidental short circuit. There was some damage to the speaker wire, which I patched by adding more layers of speaker wire on top.
This was also a bad idea because (I theorize) that over the length of about 18 inches, the flat wire was almost in direct contact with the wooden board. While wood is often thought to be a good insulator, it meant that when I left my board alone for about a month without riding it, the voltage on all 3 batteries was down to zero.
I ordered new batteries (ouchie to my wallet) and a new anti-spark switch, and re-organized my wiring so that the power switch was right next to the batteries, connected by silicone wires. There was no flat wire to wood contact when the power switch was off. This seemed to have done the trick, as I see that my batteries have kept up their charge after having been left alone for a few weeks now. Additionally, I can no longer accidentally leave my board powered on overnight (which does drain the batteries... this happened to me before. I had to buy a whole new set of the same batteries), since the new power switch I bought has an auto off after 3 minutes of no use. It also automatically turns on when the motors are moved.
The board currently weighs under 15 lb due to the downsized batteries, and still has a ~10 mile 'power-through' range (when I was really pushing the throttle, and not at all trying to conserve power).
This is definitely the single thing I spent the most time on. This is the one thing I had the most design control over, and I needed to make them perfect. I spend hours and hours on CAD making every little thing better and better until it got to a point where I decided I could print a prototype. The initial plan was to 3D print them on an SLS printer. Since you pay for mass and size, I gutted the CAD. I took out every ounce of unnecessary virtual weight from my design. That is... for both the the rear and front enclosures. I brought the cost of the prints down a few hundred dollars, then decided to print a prototype using PLA on a regular FDM printer, accessible through my school's maker lab. I printed the prototype, and both the battery casing and motor controller casing required a lot of tweaking (with a dremel). I redesigned both casings, but ended up making designs that were too complicated. School picked back up, and I just stuck with the prototype for the school year.
ESC Enclosure with tail light fitting
Battery Casing X-Ray
Battery Casing CAD
Battery Casing CAD (outer surface)
Bottom View of Enclosure CAD's
3D printed Prototype Enclosures
Previous Prototype enclosures getting ready to be replaced by the custom 3D printed enclosures
Test fitting tail light fixture on rear/ESC enclosure
Test fitting rear/ESC enclosure.
(I must have measured something wrong, as it didn't fit right. I dremeled one of the ribs to cut it down... the rib already had holes in it, which made it easier)
Charging Using Slots Cut with Dremel
After super gluing the injection molded 32T kegel pulleys from evolve to the wheels, and the super glue failing, I switched to epoxy to hold the pulleys to the wheels. This worked for some time. After a while, even the epoxy broke. I decided to switch to a whole new solution. I decided to use something weaker than epoxy. Whaaaaat? Yes, I wanted to deisng and 3D print custom pulleys. It would be cheaper than the $30 ones that evolve sells (yes, $30 each), and I could make it how ever many teeth I want it. I was not sure how well it would hold up. Since FDM 3D printed parts tend to be weak along the fused axis, I was scared that the pulley teeth might just separate from the wheel. I made up for that by incorporating M6 bolts into that axis when I designed this pulley from scratch. By strengthening the weak axis with plenty of steel, I was confident it would hold up. See the images for the first printed revision of the pulley's, printed out of ABS. (Right shows support material)
The pulleys I designed fit perfectly in the Kegel wheel cores (as I had intened). The holes were a little tight for the screws, but I was still able to thread them through and clamp down on the other into the molded plastic of the Kegel wheels (I basically used oversized bolts to thread into the plastic holes). After many miles of quick accelerations, deceleration, and even some off-roading, these pulleys are still working perfectly. The only downside is that the abs teeth seem to be slowly wearing out due to friction with the belt. This is negligible to me (again, still waiting for any signs of failure to show up), especially since the cost of material for both of these pulleys was about $2 total. I already had the steel bolts from the original kit designed to work with the ABEC core's. I would 3D print pulleys like these for future use, given they are a quick, cheap solution that works really well.
After seeing the success of my wheel-pulleys I wanted to give motor pulleys a shot. This was more of a challenge, however, because the teeth see almost 2x more revolutions than the wheel pulleys (depending on the gear ratio), and less teeth were in contact with the belt at one time. Also, I could only mount the motor pulley to the axle of the motor, meaning the contact area with respect to the applied torque was much smaller. The axle has two flat spots opposite each other. I used this to my advantage when designing these new pulleys. I made the bore on the 3D printed pulleys the exact same size and shape as the shaft. I have worked with 3D print tolerances many times before, so I was comfortable determining what size exactly would give me the right fit. I was looking for press fit that wouldn't easily slip around. I didn't want to use a set screw on the shaft because of the inherently weaker ability for plastic to hold tightly onto a screw. I came up with a design that did end up using set screws... but not in the way that one might think. I used the set screws to further tighten the already press-fit bore of the plastic surrounding the motor shaft. See the image to the right for more detail on this design.
I have no pictures of this upgrade, but it was more of a fix to a problem that shouldn't have existed. Essentially, the outer motor magnet housing/assembly started coming loose from the axle it was attached to... the main shaft of the motor. I drilled and tapped holes into the sides of the motor and through the shaft so that I should physically connect the shaft to the outer motor assembly with a bolt instead of relying on two small set screws tightened onto the shaft.
The motor mounts have come loose a number of times due to different reasons. After using copious amounts of red loctite and even some epoxy, I don't think the motor mounts will come loose any time soon.
When I re-adjusted the motor mounts, I made sure to give myself room to adjust the motor angles a little lower. This allowed me to position the motors at the perfect height to mount my trucks to the deck in the true drop-thru style it was intended to be. This gave me a board with the lowest possible profile. It is even lower to the ground (more stable) than the famous Boosted boards.
Next Steps and Why I Haven't taken them yet
Next steps for this board include the following:
shave down the width of the battery enclosure to be able to print in one piece
design better mounting system for enclosures using bolts
design work on rear enclosure to allow for better clearance around motor wires
incorporate taillights and headlights (already designed)
incorporate telemetry package (have components)
order bigger wheels (85mm caguama)
add nighttime reflectors
Why have I not done any of this? Well, I decided that I wanted to build an entirely new board. One that was going to be built better in every way that the original board wasn't. After this new board is built to satisfaction, I will go back and upgrade the old one to use around campus, and when I need a lighter 'safe bet' board. Turns out my old board is also just much more fun to ride. It's more carvey, as the other one was built more for speed than carving.