Weight Sensing Longboard with Inline Wheel Motors

You have been posting a lot about eboard stuff lately ...it seems as though the your eboard fever is rising again Justin!.... its great to see you getting on-board ... may i ask what has focused your attention on the realm of eboards lately?

....ebikes are so 2014... the future is eboards?
 
justin_le said:
Weather permitting I'll ride it again back to work in the morning, this time taking the spirit trail to the lions gate bridge and then riding through Stanley park and along the waterfront bike path back to the shop. Should be a thrilling way to start the day.
Click to expand...

And indeed it was. Here's the whole 30+km commute loop:
FullCommuteMap.jpg

The lion's gate bridge is a tad nervewracking on a powerboard. The joints between the bridge segments are harsh on skateboard wheels, and the railing on the mid-section leaves a gap that unlike on the 2nd narrows bridge is large enough that the board could shoot right through it. So the risk of bailing and then watching the longboard roll overboard is a definite possibility
LionsGateFit.jpg

Before I left in the morning I also reduced the CA data's averaging so that we could look a bit more closely at the signal dynamics on the board. The special skateboard CA firmware has an 11Hz data output rate, so it gives good time resolution of all the weight signals on the trucks while riding. In the CA trip analyzer program, the front weight shows up in the "H.RPM" column, and the rear weight is in the "H.Watts" column, purple and green lines respectively in the graph snipped below, every time there is a bump in the track you see these spike up to 200+ lb.
BridgeBumpSignals.jpg
http://www.ebikes.ca/tools/trip-analyzer.html?trip=zWNULA

By comparison the ride down stanley park drive was almost completely smooth, although it is surprising to still see the extent of noise on the strain sensors from the front and rear trucks.
StanleyParkDrive.jpg

This doesn't cause any power ripple on the motors because we have a ramp rate limit set on how quickly the CA can increase or decrease the torque command on the controllers, but it means that if we wanted to do control of the motors to detect when someone has their foot off the deck and is kick assisting, then there will be some signal conditioning required to pull that out of all the noise.

There were a few glitches towards the end of the ride in to work. In an effort to avoid a patch of rough brickwork I ended up riding straight through a large puddle of water. My electronics is mostly exposed and some drops splashed on the CA board. Didn't mess up the weight signals luckily, but it did get on the voltage divider and so the CA thought that the pack was at 45-50 volts all of a sudden. As a result of this the low voltage rollback didn't kick in, and I ended up tripping the BMS low voltage cutoff circuit at high loads just before making it into the shop. For doing this on a regular basis, I'd certainly like to try larger wheels and/or more suspension with a higher flex deck.
 
onloop said:
... may i ask what has focused your attention on the realm of eboards lately?
Click to expand...

Oh I've been waiting for this moment for over a decade, it's all finally coming around full circle which is great. When we started the UBC ebike club in 2003/2004 about half the projects were electric boards, starting with the loaner ExKate that I started this thread with.


Plus a few 3-wheeler board using small scooter hub motors. These were geared motors and couldn't do regen, so we had a cable actuated drum brake for slowing down:
UBC 3 Wheeler.jpg
View attachment 2

And then the balancing 2-wheeled "project emanual", which is definitely due for a 2015 build update
View attachment 1
If anyone's interested I saved most of the website we had for that and put it here:
http://www.ebikes.ca/emanual
Though all the original videos were hosted on the long gone google video.

All this while I've been telling anyone who will listen that while sure, ebikes are a great enhancement to regular bikes, electric boards are totally transformative and could create a whole new class and category of vehicle that the world hasn't really seen yet. It's really cool to see this start happening in the past few years with so many startup businesses bringing well made complete eboards to market, and also to see so many people here embrace it with the same DIY vigor that characterized the north american adoption of ebikes too.

So why now and not 10 years ago? Pragmatically ebikes are a much easier sell since they fit inside an accept transportation norm and regulatory framework, but mostly it's just that running an early stage ebike business has been so %!!*%#$% time consuming that it would have been foolish to try and carry on all the board experiment with everything else on the plate. But ebike tech has matured and gotten more streamlined the last little while which has freed up some R&D time, and seeing the buzz around boards on this forum has gotten me really excited to fulfill building the dream e-boards that we had sketched out back then (inline hub motors and weight sensing control).

....ebikes are so 2014... the future is eboards?
Click to expand...

:D For some people for sure. I won't go so far as that as a generalization; the future for inner city transportation should include human sized EV's of all kinds. Boards for some, trikes for others, bikes for many, and giant self driving cars? BLAH, take that american idea of transportation utopia right back to the suburbs.

My guess would be weather. First weeks of sunshine in the PNW.
Click to expand...
That's part of it too!
 
I've been practicing riding a board, and it feels comfortable now.

Sign me up for a board with your clever weight shifting electronics for control and direct drive hubmotors. My area is extremely skating friendly, it's not uncommon to see guys skating to work in a suit and tie in the mornings in Santa Cruz.
 
justin_le said:
If anyone's interested I saved most of the website we had for that and put it here:
http://www.ebikes.ca/emanual
Though all the original videos were hosted on the long gone google video.
Click to expand...


That's some history right there!!

I don't remember seeing this one,
cool!

Board_Playa_Small.jpg
 
sk8norcal said:
I don't remember seeing this one,
cool!
Click to expand...

That's actually been the most successful of eboards I've build so far in terms of years of continuous usage, build in 2006 for burning man, here it is at the 2008 car free festival (the remote in the hand is just for steering, there was something up with the differential lean sensors at the time):
View attachment 2

Upgraded the deck to use the ends of a snowboard rather than plywood to good effect and had it at the several Maker Fairs right up until 2012. The sharp snowboard edge did gouge a few shins so eventually I put a rubber bumper on the nose and tail:

Snowboard Emanual Maker Fair 2010.jpgSnowboard Emanual Maker Fair 2010, kids.jpg

The large wheels made it super easy to ride at low speeds, but I found high speed travel on it much more of a challenge than with the original Emanual board using 4" caster wheels.
 
Last year Lone_Deranger mentioned the Rojas Hybrid trucks on this thread,
http://endless-sphere.com/forums/viewtopic.php?p=904580#p904580
Rojas Product Image.jpg

Up until this point I had always been chucking the truck hangar in a lathe in order to machine a round section concentric with the axle for the motor to slide on, which isn't something that most people have the ability to do. But the Rojas trucks have a totally round steel hanger that means we can fit an inline motor on the truck with no modification. We contacted Braden Boards last summer to see if they'd be OK with us trying this and indeed they were great about it and so we ordered a number of samples, which arrived promptly but then sat in a box :(

That was until a couple weeks ago, when it was finally time to bust these out for our 4th inline motor weight sensing longboard build. The hangar is about 11.5mm diameter, while we had our motor cores machined for a 15mm ID. So the motors we made slide on with a TON of room:
Rojas Axle Fit.jpg

While it will be easy to machine a shim sleeve to fit between them, I'm actually growing on the idea of not rigidly mounting the motor to the truck hangar, but instead letting the motor 'float'. So it would be tortionally coupled between the wheel and the truck, but it's axial location wouldn't be rigidly defined. That way we could ensure that the motor bearings don't see any kind of harsh impact loads from riding over rocks and hitting bumps, those forces would be 100% taken by the wheel bearings. And if the motor did take an impact, it would mostly deflect out of the way.

I'm not sure if it's strictly necessary, but the fatigue wear and vibration exposure of the motor is one of the things that worries me the most for long term reliability of the inline wheel motors, especially given the thin section ball bearings that are used.

Anyways I went ahead and did the strain gauge installation on one pair of the sample trucks. The first one was done exactly as the others, with a pair of gauges on the top and bottom so that we have a full bridge strain configuration with all 4 gauges active for maximum sensitivity.
Rojas Top Strain.jpg
Rojas Bottom Strain.jpg

But on the 2nd truck I put all 4 gauges in a single cavity on the bottom, with two gauges that will be in tension, and the other two are mounted perpendicular to the strain so should just be passive resistors. This should have only half the signal intensity of the full bridge with all sensors active, but if it's good enough then it would be much simpler to put together. We could get just a single strain gauge manufactured that has all 4 strain sensors etched on it and connected into a bridge, and there would be just 1 flat spot to machine on the truck, 1 gauge to glue down, and then 1 cable to attach.
View attachment 1
 
excuse my ignorance but how exactly does the strain sensor work? what is it actually measuring?

but before you tell me let me have a guess.
umm... so my guess is.... that the when increased load is put onto a lever the surfaces at the fulcrum stretches or distorts somehow, and somehow this device is measuring microscopic surface movements....

was i close?
 
onloop said:
was i close?
Click to expand...

Not far off. Strain gauges are just resistors etched in a zig zag grid, and when the surface of the metal is under tension, it stretches, elongating and thinning the resistor wire causing its resistance to increase tinyest amount (<<1%). Similarly, when the metal surface is under compression, the resistive wire gets shorter and fatter, lowering its resistance by a fraction of a percent. It's one of the most deceptively simple and ridiculously accurate electromechanical sensor devices, but it does require a high precision low drift instrumentation amplifier to turn the bridge output into a 0-5V signal.

I already detailed the amplifier circuitboard and strain gauge operation on this thread here:
http://endless-sphere.com/forums/viewtopic.php?p=904058#p904058

I got most of my strain gauge education from Richard Nakka's rocketry website way back in the day. 15 years later his website is still there and still a great reference:
http://www.nakka-rocketry.net/strainlc.html
 
Very cool... I'd like to try you board.

Riding without a controller would be interesting. As long as it is super responsive and you can stop just as quick as when you use a trigger system.

How about making a module, that has the sensors and amp, that works with other esc. I suppose it needs fairly advanced esc... Such as vesc..?

What interface is required to connect with esc.

Is this technology likely to be something that is plug an play any time soon?
 
So I read through that rocket site you linked, most of it I skimmed over because I didn't understand.

But I get the gist of it.

So here is my question, and of course this is that exact opposite of what you are building (skateboard without remote) , is it possible to make a wireless hand control with zero moving parts?

So basically the rider is holding a piece of aluminium with a shape that is comfortable for the hand to grasp.

It could have a stationary/fixed trigger for index fingers. Or all fingers.

But basically it senses when you squeeze the 'fixed' trigger. Measures the strain. Transmit signal to receiver as usual?

Would it make sense to replace a potentiometer with strain gauges?

It would make for a really cool looking remote.
 
onloop said:
So here is my question, and of course this is that exact opposite of what you are building (skateboard without remote) , is it possible to make a wireless hand control with zero moving parts?
Click to expand...

Yes of course. I've already experimented with this approach as a throttle for ebikes and Arlo built a very nice regen brake lever for a motorcycle; the lever is fixed but a strain gauge senses how hard you are squeezing and it works brilliant:
http://endless-sphere.com/forums/viewtopic.php?f=1&t=61397&p=918073

Would it make sense to replace a potentiometer with strain gauges?
Click to expand...
Yes and no. It would be much more reliable in the long term than a pot, which will inevitably develop wiper problems. But it's also way overkill for precision and accuracy than what you need in a hand squeeze sensitive throttle device. There are other ways to sense squeezing force that are more than adequate for a throttle and a lot cheaper for mass production. A flexure with a moving magnet and hall sensor is one common approach, and force sensitive resistors are really simple to use for this kind of thing too.

It would make for a really cool looking remote.
Click to expand...
Then don't wait for it to happen, go out and build one!

-Justin
 
Jonesing for an update on this Justin. Any work on version 4 lately?

justin_le said:
Up until this point I had always been chucking the truck hangar in a lathe in order to machine a round section concentric with the axle for the motor to slide on, which isn't something that most people have the ability to do. But the Rojas trucks have a totally round steel hanger that means we can fit an inline motor on the truck with no modification.
Click to expand...

Clearly your plan is to make a kit that people can buy and combine with some Rojas trucks. Are you thinking an in-wheel motor kit or a weight-sensing board kit or perhaps both? Is that a long term plan or a medium term plan? Basically what I'm trying to say here is "take my money!"

Will you be bringing version 4 to the Vancouver Maker Faire?
 
What Agraham said "take my money"! Already have the needed trucks, and will willingly set the lathe loose on my 107 flywheels!

On a side note, your going to love riding those rojas! :D
 
Yea this thread makes me want to start practicing on a skate board again.

As a kid I used a skateboard to get around and worked quite well. Just been a long time.
 
ohzee said:
Yea this thread makes me want to start practicing on a skate board again.

As a kid I used a skateboard to get around and worked quite well. Just been a long time.
Click to expand...

Same here! Except that I do live close enough to work that I could ride a board there, and say, enjoy a cigar on my way. Not to mention it would be perhaps the very fastest form of on-campus transportation. Our city is very hilly (I ride an 18%er every day) so regular boards aren't practical. You'd be walking 60% of the time, uphill or downhill.
 
Nordle said:
Any chance to buy a set of strain gauge + pcb? how does it work, pwm?
Click to expand...
Google wheatstone bridge.
 
You must log in or register to reply here.

Similar threads

O
ITC3 - Balance-controlled skateboard project
Replies
2
Views
892
Flexscoot
F
justin_le
Grin's weight sensing electric longboards #4 and 5, and kit!
2
Replies
27
Views
11,003
lone_deranger
L
CoolRextreme
My Second Eboard Build; 12S Dual-Drive Longboard
Replies
3
Views
2,004
CoolRextreme
CoolRextreme
C
Struggling with choosing an electric longboard for commute
Replies
15
Views
2,912
torqueboards
T
sk8norcal
Superfast Skate (Groundsurf)
Replies
10
Views
4,354
sk8norcal
sk8norcal
Back
Top