Weight Sensing Longboard with Inline Wheel Motors

[melodious]

My gut instinct on the motor "in" the truck/axle says it won't last long. The anchor points need to be beefier. Give it a good pounding and prove me wrong.

I guess there is no way to further reinforce the motor via welds?

 

[justin_le]

Text to follow after sleep:



View attachment 3

 

[Hummina Shadeeba]

The motor/wheels are awesome. The simplicity appeals to me. If they could work on regular trucks the world will go ape. How are you getting the motor to stay in the wheel and any plans to make it removable or you just going to cut it out?

The weight sensing is interesting but I think your hands are probably the best place to keep control..not many people are doing well with foot-controlled video games.
The hole for the screen..leaves a big hole in your board. Carbon it up!

 

[LEVer]

The weight sensing is interesting but I think your hands are probably the best place to keep control..

This is not like a Zboard, its way better. There is no lifting of the feet to control acceleration and stopping. It's even better than mine, which I know is better than the Zboard. The only differences are that I have mechanical brakes and freewheel clutches. But for the most intuitive and the ability to be the stealthiest-looking longboard, this wins by a mile.

Justin, I have questions about your strain gauges. The way I understand it is that they are measuring the strain on the hanger/axles of the trucks; would this not give a false reading based on the smoothness (or lack thereof) of the road surface instead of the input of the rider? Also, I'm assuming it does not go in reverse and you mount the board stepping on top of the rear truck first, right? Otherwise, the board will move in reverse or you would have to mount on the board with both feet simultaneously. Why not just put Force Sensitive Resistors on the risers?

I can see this setup and interface, add in the ability to control the parameters via a phone app, to be the next e-longboard design best suited for urban travel. Great work.

 

[Hummina Shadeeba]

"This is not like a Zboard, its way better. There is no lifting of the feet to control acceleration and stopping. It's even better than mine, which I know is better than the Zboard. The only differences are that I have mechanical brakes and freewheel clutches. But for the most intuitive and the ability to be the stealthiest-looking longboard, this wins by a mile."

i think i rode the zboard and found it not surprisingly awful. Even the guy who let me try it didn't like the acceleration control. Trying to balance on a skateboard with your feet and also control acceleration and braking seems to me to be overwhelming no matter how well programmed. Maybe I'm missing something important but it sounds like playing grand theft auto with just your toes.
I haven't seen or tried or really know anything about this board or your board though and I don't know much about electronics..but I cant imagine it not having a huge downside in board manuverability. I like the intention to integrate it with pushing and I think it could be an effective seamless integration and have ti's moments but...I dont want to push at all. Maybe I'll need both the wheels and this tech to sneak by the law if they keep up the ban in cali. the ultimate stealth vehicle as you say.
those wheels though are great. I wonder how fast they could go on 6s.

I'm trying to make a motor mount made for 44mm motor bolt holes since I can't find one. It's seemingly going to cost a fortune and what I really want to do is get rid of the mount, pulleys, n belt and make this.

 

[LEVer]

I agree it has a learning curve, but there's something about using your hands to e-board. To me, it's just like an e-scooter with invisible handle bars, just IMO.

The beauty of this is that you "do not" use your feet for acceleration and stopping, you just shift your weight. It's a "true" weight-sensing eboard. It's like a snowboarder with invisible foot-straps tearing the flatlands. If you want to stop, just lean backwards.

Anyway, a simple phone app can be added to disable the weight-sensors and use the phone as the remote-control. You can even use the accelerometers in the phone to control the board.

 

[kfong]

This is a well thought out design, Justin. Thanks for sharing! Can't wait to hear performance figures with the smaller motors. If they are good, I plan to model my longboard after yours. We just CNCed our lathe this past month, it's just waiting for projects like this.

 

[justin_le]

The motor/wheels are awesome. The simplicity appeals to me. If they could work on regular trucks the world will go ape.

These are regular Randall trucks, nothing fancy about them other than turning down a constant diameter on the ends to give a concentric support for the motor, and of course gluing on the strain gauges. Agreed it would be super awesome to be able to fit the inline motors without any mods to the truck, but the nonstandard profiles of truck hangers complicates things. We came up with a few schemes for this where effectively the motor is supported by the 8mm steel axle instead of by the diecast truck hanger, and which then presents a 10mm axle for the skateboard wheel bearings.

[How are you getting the motor to stay in the wheel and any plans to make it removable or you just going to cut it out?[/quote]

Everything assembles and disassembles quite easily, no need to cut anything out. The motor just press fits into the rubber wheel which was bored out to have a slightly smaller in diameter than the OD of the motor, so it holds well with no slippage, but pulls apart by hand without much fuss. Have a close look at the section view, you can see that there is interference between the motor and the bored portion of the wheel, but the rubber wheel is soft enough it stretches over fine and holds it tight

The weight sensing is interesting but I think your hands are probably the best place to keep control..not many people are doing well with foot-controlled video games.

This doesn't have much to do with foot controlled video games, as others have pointed out the analogy of surfing and snowboarding is much more appropriate. Riding an electric skateboard with a joystick control is like riding a surfboard with a hand linkage to a rudder for steering the board. My initial motivation for this project came from riding an original eXskate board and thinking that hands are totally the wrong place to control a skateboard. You really had to train yourself to lean before you pushed the trigger and coordinate all the finger motion and body weight shifting, the brain figures it out like anything but it's not natural.

The hole for the screen..leaves a big hole in your board. Carbon it up!

Yeah, if there's ever going to be a deck failure we know where that will be at. But having the display screen inlayed is so cool.

 

[justin_le]

Well guys, been riding around the block all night and we can't get off this board it is so much fun.

For the battery pack, we assembled a total of 3 of these 10Sx1P batteries from 2600 mAh cells, for a total of 37V 7.8Ah. We made the plastic battery enclosures with an open wire channel that spanned their full length so that we could run all the cables between the front and back of the board out of sight. One thing we learned from the first build was that cable managagement on the under side of a deck can be a real mess, especially with dual motor controllers, strain sensors, parallel connected batteries etc. and hoped to be able to make this one clean from the get go. There are also end-to-end andersons embedded in the housing so that the packs could all plug in in parallel but with the connectors out of sight:


We wanted a long flat battery in order to minimize the side profile of the pack, but the downside of a single large battery is that a board deck flexes quite a bit, and batteries with all their tab welds etc. don't take to repeated bending very well. So by breaking it up into 3 discrete packs the total bending displacement per battery is a lot less. Each pack enclosure has it's own effective stand-off on either end where it is screwed to the deck, and supports the battery with about 3mm of clearance to the wood


This way there is allowance for the wooden deck to flex a fair bit without causing any bending forces on each of the batteries. And jumping on the deck seems to confirm this. There is no loss in board flex from having the packs, and the batteries stay nice and straight while the board springs down



For the controllers, we again used the ASI BAC500 devices controlled by a Cycle Analyst. Once tuned to the motor parameters these run sensorless flawlessly, with proportional forwards and braking torque control of the motor, and with total smoothness and silence. Again to keep the wiring clean, we had both controllers paired to behave like a single dual controller, with only one battery connection, one Cycle Analyst connection, and then the two motor outputs:


The Cycle Analyst PCB and strain gauge amplifier boards I have coupled together at the front of the deck, covered for now with a plate of lexan in order to show people the circuitry at Maker Faire, but long term we will protect this with a proper enclosure too:


On the finished board, you really need to be down low at an angle to see any of the electrics, and the only hint from a regular vantage point of the power inside is the faint screen glow emanating from the CA

 

[lone_deranger]

Although more expensive than Randals (but cheap for precisions) this truck design may require less machining.

http://bradenboards.com/products/trucks/
A concern with modifying a cast hanger is a loss in the structural integrity - they do have a tendency to banana through normal wear and tear.

 

[justin_le]

Although more expensive than Randals (but cheap for precisions) this truck design may require less machining.

http://bradenboards.com/products/trucks/
A concern with modifying a cast hanger is a loss in the structural integrity - they do have a tendency to banana through normal wear and tear.

Dang, those would have been Ideal for this project. Absolutely ideal. I will order a set for sure for the next prototype, since if we can design an inline skateboard motor that at least works with one model of off-the-shelf truck out there, it would still make it way more accessible to all the people without machine shop tools. Thanks for sharing that.

 

[justin_le]

Justin, I have questions about your strain gauges. The way I understand it is that they are measuring the strain on the hanger/axles of the trucks; would this not give a false reading based on the smoothness (or lack thereof) of the road surface instead of the input of the rider?

In almost all strain gauge circuitry you are running a low pass filter in the amplifier, so any high frequency changes in the strain (as you have indeed on a skateboard truck) are filtered out, and only the average weight remains, unaffected by all the rapid coming and going of bumps in the road. So no there aren't any false readings, and even if there were, they would still have no net contribution since vibration noise averages out to zero.

Also, I'm assuming it does not go in reverse and you mount the board stepping on top of the rear truck first, right? Otherwise, the board will move in reverse or you would have to mount on the board with both feet simultaneously.

The Cycle Analyst is configured so that it doesn't command any power to the controllers until the board is actually moving faster than about 2mph. So when you just stand on it, nothing happens, give it one kick and off you go.

Why not just put Force Sensitive Resistors on the risers?

It's tempting, and my friend Frank who worked on this many years ago (see start of thread) did his first builds with pressure sensitive resistors. But they aren't nearly as accurate or as linear as strain gauges. You can deal with non-linearity in software, but they aren't very stable or consistent either. I'm sure you could design a pressure sensitive riser that simply installed under the truck and would be a bit more friendly for most people to use, but since I have strain sensors and circuits on hand from other projects this approach was easiest for me.

For later firmware enhancements I want to try and have the electronics detect the weight shifts that happen when you are kicking and then change the control behavior, so a sensitive and accurate measurement of the weight would be important for that. And as a plus, it actually makes a reasonable accurate digital scale.

I can see this setup and interface, add in the ability to control the parameters via a phone app, to be the next e-longboard design best suited for urban travel. Great work.

Thanks, I'll be happy to see more e-longboards of any kind making inroads in urban travel. That would be a future I could live for, so everyone else doing or considering electric longboard builds, Go for it!

 

[kfong]

Is there software we have to download to use the CA as a skateboard controller. Justin, will you be making that available. I'm sold on making a similar setup. Which CA do we need to buy?

 

[LEVer]

Justin, thanks for your reply.

I think this truck would be easier to mount the motor on. You can take out the shaft of the motor and replace it with the replaceable 8mm axle. You can also use the other tapped hole to bolt and secure the motor on.

 

[Hummina Shadeeba]

Justin, thanks for your reply.

I think this truck would be easier to mount the motor on. You can take out the shaft of the motor and replace it with the replaceable 8mm axle. You can also use the other tapped hole to bolt and secure the motor on.

those trucks were what I was looking for to possibly cad print a truck hanger with a mount attached and then screw in the axles. Do u know if u can screw those axles into solely aluminum.. Don't know if u can do that. ( Got someone to make some I'm going to weld on but it took forever.) Then again I don't really know if that could work but I'd be a great freedom if you could
Just print the hanger and get some axles that'd screw in.

I still can't imagine pressure sensing boards being as controllable. What if you were to slip and lean toward the front or back...because you're going 25 on a skateboard, and then you're further jostled by the changes in speed from your foot weighing somewhere just trying to balance yourself?

 

[sk8norcal]

I still can't imagine pressure sensing boards being as controllable. What if you were to slip and lean toward the front or back...because you're going 25 on a skateboard, and then you're further jostled by the changes in speed from your foot weighing somewhere just trying to balance yourself?

had a chance to ride Justin's board at Maker's faire, there is a delay in the response to prevent the scenario ur describing.

One thing I was impressed with was that his board was very smooth in acceleration from stop,
(the pushing feature was not available at the time, there were some funny awkward bails when some people tried to footbrake :lol: )

overall its interesting, but I wasn't really convinced that weight-sensing is the way to go...

 

[LEVer]

This debate seems to mirror the "hand-throttle vs pedal-assist" in e-bikes. Why not have both?

 

[torqueboards]

That's such a brilliant idea with those battery packs and having them bend in those correct places... awesome work!

What did you use for your battery enclosure? Did you CNC mill the enclosure out of a plastic block?

 

[ZBoardBen]

One benefit of weight-sensing boards is the ability to pick it up and still have a hand free (to open doors, your trunk, etc). This is becoming less of an issue now that controllers are getting smaller, but back when they were bigger you couldn't put them in your pocket easily.

Also its nice to have both of your hands free when riding if you need to carry something large. Lastly you don't need to keep track of your controller, or keep it charged. Small benefits for sure but they add up.

Justin your build looks awesome btw. I'm curious, what would happen if you tried to foot brake? Also, if you theoretically kept your weight distribution completely consistent, and then went into a tight turn, would the board accelerate? That was an issue we had with this one of our original prototypes (http://goo.gl/IEA6Rl) which led us to having discrete sensing areas.

Note - we recognize that our weight-sensing control scheme takes some learning, which is why we offer a 30 day guarantee. If you can't learn to ride (or you don't love it, wasn't what you were expecting, etc) you can send it back for a refund.

 

[melodious]

@weight sensing, I think it's the only step to innovation. If the board had a handheld transmitter, it would be similar to 99% of the Eboards here. The reality is that the body's position is leaning or lunging forward into & infront of the forward trucks. This establishes more precise control for steering at speed.

Personally, I'm not sure my style of riding is suited for this type of trigger. I stance dance all over the board, and I'm wondering what effect the distribution of all the various weight shifting would do to the ride

 

[bandaro]

Congratulations justin, you have proved me (and many others) wrong or settled our fears and/or curiosities. An incredible build all around, making me reconsider plans on a board design; "hub motors" for skateboards may be the way.

So now I'm curious as to it's freeride abilities? It's all well and good to use one for a commute along a road, so how does it survive the sliding and rapid weight shifts when you jump, ollie, ride off curbs, etc? Does the delay in controls allow time for you to slide/land/whatever and continue normally? I'm throwing about some design ideas for a hand controller that is both small and compatible with slide gloves, as I feel that is more in my league than the controls for weight sensing.

Can you slide at all or will the wear on the wheels be too great and wear down to the motor too quickly? Just working on some random musings for build options...

 

[justin_le]

What did you use for your battery enclosure? Did you CNC mill the enclosure out of a plastic block?

Yup, I bought a 2' x 2' sheet of 1" thick UHMW and we hogged them out of that. It's not the most cost efficient, we only got 10 pieces out of the $86 plastic sheet, but it means we could viably do a small production run of e-board battery packs which any one can just screw onto their deck like this. The stuff is slippery. I'm not sure if it'd really be wise to do a rail grind on the belly of this board but if you did you'd surely slide pretty fast!

 

[justin_le]

Congratulations justin, you have proved me (and many others) wrong or settled our fears and/or curiosities. An incredible build all around, making me reconsider plans on a board design; "hub motors" for skateboards may be the way.

Well at the very least it should be an option on the table that people have. The belt drive will always be able to get more torque for the same size motor due to the advantages of the gear reduction, but direct drive should have the lowest rolling friction when used as kick board board without power. Both requires some mods to the drive wheel itself, but hopefully those mods are easily doable at home with minimal tools since wheels wear down and need replacement.

So now I'm curious as to it's freeride abilities? It's all well and good to use one for a commute along a road, so how does it survive the sliding and rapid weight shifts when you jump, ollie, ride off curbs, etc? Does the delay in controls allow time for you to slide/land/whatever and continue normally? Can you slide at all or will the wear on the wheels be too great and wear down to the motor too quickly?

Unfortunately I am the wrong guy to answer that, there was a time in the 90's when I could land a kickflip but that's about it, right now slides and jumps etc. are not in my ballgame. This is why I'd really like to have some more experienced longboarders try this out for their kind of feedback, not just around the block but on some long trips and descents. When you are really used to a certain vehicle behavior it takes a longer time to get used to and familiar with the nuances of something different, so for a fair comparison they'd really need to ride it for quite a while.

For NON experienced longboarders the weight shift come completely natural and the learning curve is just a few minutes. We had hundreds of people test riding it over the weekend at the maker faire, from 10 year old kids up to some grandparents, and most of them were pretty hooked. Learning to kick and maneuver a regular longboard is a way more challenging task.

 

[justin_le]

One benefit of weight-sensing boards is the ability to pick it up and still have a hand free (to open doors, your trunk, etc). This is becoming less of an issue now that controllers are getting smaller, but back when they were bigger you couldn't put them in your pocket easily.
Also its nice to have both of your hands free when riding if you need to carry something large. Lastly you don't need to keep track of your controller, or keep it charged. Small benefits for sure but they add up.

I'm with Ben on all these points. Hands free is such a ticket to do rad stuff while riding the board.

Justin your build looks awesome btw. I'm curious, what would happen if you tried to foot brake?

If you tried to foot brake with your back foot, then the board accelerates and you quickly learn that's not a good idea. If you foot brake with the front foot, then you come to a stop even faster. We kept telling people at the faire to lean back until they come to a stop and then get off. Most people had no problem with that, but those who were boarders would regardless try to foot brake and run into one of those two issues until it was trained out of them.

Also, if you theoretically kept your weight distribution completely consistent, and then went into a tight turn, would the board accelerate? That was an issue we had with this one of our original prototypes (http://goo.gl/IEA6Rl) which led us to having discrete sensing areas.

That's a good question. Theoretically if your weight is even on both trucks then there is no acceleration, since it's just looking at the force differential. But on the field it's for sure trickier to keep your weight balance in check when turning sharp corners; it takes a bit of practice to pull a 'U' turn on the road and keep your speed consistent the whole while. I tend to lean forwards when turning inwards, and lean backwards when turning outwards, so in one case I tend to accelerate a bit unintentionally and in the other I tend to decelerate. Once I notice what's going on I adjust, but it's still not perfectly smooth all the time.

The way that I have the gauges mounted on the trucks, I could actually wire each truck up as two independent half bridges instead of a single full bridge, and then sense how much leaning was going on the hangar and possibly adjust the controls when you are cornering. Might be quite interesting to do that for R&D related stuff but I don't think it's necessary for a viable control scheme.

The old prototype build looks pretty cool too, btw. Do you have any more build images or details on that?

 

[liveforphysics]

I'm perhaps the worst skateboarder who is still willing to try it occasionally (typically wiping out within <10ft). I had the treat of getting to very cautiously try Justins board, and had not only the fastest smoothest coolest skateboard experience of my life (very surfing like), but it was my first skateboard experience that didn't end in wiping out,

The hill climbing torque on this board was incredible. The climb to my home has a section of >20% grade hill, which is quite the treat for me to climb on my roadbike. His skateboard just hammered up it at an impressive rate. His skateboard is also capable of drawing some serious current.

If you want highest performance, it's inherently going to be something with as little non-active material as possible. That means direct drive hubmotor.