Project: MonoWheel (Experiencing the physics of a forgotten invention by re-inventing it)

1. This is is a little docu of my MonoWheel build. The idea sparked on one of my recent Unicycle rides , which is typically pretty stressfull as it takes me so much attention and control - when I just want to enjoy the roll- and tilt degree-of freedom of a one-wheeled vehicle while crusing with the family. However, the uni-ride that day ended with a minor crash, while taking over my kids and my shoe-laces getting stuck in the crank (Ouchhh).

   

   Fig. 1 : Stuck in the unicycle crank

2. Having gone through other types of unstable vehicles prior, I really wanted to go for sth. on one wheel, but inherently easier to balance by a human @ relatively low speeds (unicycle speeds is perfectly fine for the type of riding I do). Back in the 1860's - the time of the original invention of the MonoWheel, a main driver for the size of a wheel was certainly driven by factors like absece of smooth road surafces , and a large wheel radius was simply reducing the rolling-friction to some vehicle on land (including chassy for early planes ). There was certainly not much concern about moment of internia of a wheel. Since the moment of internia scales the the square of the radius (J~r^2), big wheels tend to become quickly very difficult to steer , and which is one important reason for its lack of gaining popularity for main stream transportation . Here I was intentionally looking foward to use this extra rotational inertia to control the bike through the torque vector on the grounded gyroscope (the special case of Leonard Euler's fundamental laws of motion for rotating solids). I saw myself prospectively experiencing the same physics in a new way - as we do , controlling a Helicopter, scrubbing a motocross bike (to lower flight height by transforming linear motion into rotary motion during flight) or laying in the hospital (eg after some MX crash..) and trying recall how these beautiful images of your brain on the screen of the MRI sytam are generated (namely, the magnetic moment of the spinnig nucleus within an atom will try to re-align, interacting with an external magnetic field; The torque required to alingn this rotor will cause the nucleus perfroming a wobbling motion (gyroscopic precision) at a characteristic frequency, which is being analyzed through spectroscopic methods ) . Feeling this in the macro-/real-life world is sth. what always gets me excited, and why I got started to tinker the technical stuff after defining the physics and its boundary conditions.
3. So, how to get wheel made - as large as possble, yet to potentially still carryable on a car w/o trailer, light enough to propell it purely by muscle power (maybe some electric motor assistance at best), stable enough to sustain the pressure of the rollers on the hubless wheel as well as stable enough in plane, in order to sustain warping during the potentially extremely quick turns of the wheel . Of course this has to be done with easily accessible materials and tools.
4. I decided to build the 2 m dia. wheel from ABS, which was readly available in handy dimensions at Homedepot & vey cheap. All started by cutting a first set of 1.5 in ABS tubelets of 40 mm height on the mitre saw

   

   

   Fig. 4 : Making the inner ABS layer

5. The next task was to build a fixture , which will center the wheel during the buildup from the inner -to-outer. Some funiture boards did the job, but I had to make this template very precisely round with ca. 80 cm radius. The home-made circle was comprised of an old Honda 500 connecting rod , a piece of wood , a pen and some means to adjust, so that the whole thing remains stable (Good moments to discuss geometry & improvisation with the daughter...).

   

   

   

   

   Fig. 5 : Making the core circle template

6. Next we fused the tubelets of the fisrt layer together .

   

   

   Fig. 6 : The inner ring is done

7. So did we make the second layer.

   

   

   Fig. 7 : Second layer

8. Since we are dealing with a curved substrate, our closest packed structure will contain gaps between tubelets of the second layer & onwards. Since these permeter links are essential for wheel stability, I decided to fill the gaps , and installed so, that we have only compressive stress along the entire perimeter under each loading condition (front-back/up/down / with rider payload & reserve loading). This technology worked pretty well , and brought the expected stiffening after the 2nd round alredy.

   

   

   

   

   

   Fig. 8 : Introducing the intrinsic strain

9. This is the second weekend I am starting to work on the wheel. I am beginning with installation of the 3rd layer

   

   

   

   Fig. 9 : third layer competed

10. and the 4th layer. (the spacer cutting was quite dangerous on the electric mitre saw,and since the reciprocting saw was slow I finally was fastest by cutting them by hand saw). Another upright check proofs that I am on track with stiffening goals.

    

    

    

    

    

    

    

    

    Fig. 10 : Fourth layer is now done

11. The 5th layer gets easier , since the next standard ABS tube size (2 in) fits just on the 4th row, compressed, but w/o the spacers.

    

    

    

    Fig. 11 : Thie final row is installed

12. Now we start building the veneer layer all around the base honey-comb-like structure (6-nodes per unit cell) to maximize the structural stiffness. This layer will connect all remaining open nodes with each other . Starting with the outer permeter veneer.

    

    

    

    Fig. 12 : installing outer perimenter veneer

13. We repeat the process of node-bonding on the inner rim

    

    

    

    Fig. 13 : installing inner perimenter veneer

14. The Wheel is now stripped off the remaining paper residues and then planarized with a belt-sander from both sides. The Wheel has now almost the final radial stiffness , and weights just above 13 kg @ slightly over 2 m dia.

    

    

    Fig. 14 : Cleaanup and planarizing

15. The axial veneer layer is being installed now. Cutting ABS sheets with jigsaw, applying the adhesive with paint roller. Quickly loading the surface with multiple deadweights of all sorts (Yard stone, Cylinder barrels, Flywheels) in order to maximize bonding uniformity

    

    

    

    

    

    

    

    

    

    

    

    

    Fig. 15 : Installing Axial veneers

16. the Third weekend begins. Tinkering the inner wheel (the hub) after beer on friday night . Planned to coninue in ABS, but quickly realized that all drive mechnism I need as well (some kids bikes) would not work well together in terms of joining & overall stiffness.

    

    

    

    Fig. 16 : Friday-night concepts after beer

17. What finally did the job was the lumber I had laying around from the last project on the house. Cutting out the wheel housings with cicular-saw , and installing the skateboard rollers via 8 mm spindles

    

    

    

    

    Fig. 17 : Installin the radial rollers

18. After setting up the lower radial support (The Bike wheel), & alignment of frame to ensure that the clearance of front an back wheel was sufficient (>5mm) was in place (the hypothesis is that themain load sits on lower wheel, and that too much load front /back wheels may overload the MonWheel / potentially let the structure fail by fracture of nodes ), I instlled the axial rollers Front&Back (daughters roller skates pair had to suffer here).

    

    

    Fig. 18 : Installin some radials and axial bearings

19. Saturday I installed the lower Axial bearings on wooden subframe as well as made the upper roller support . The first test ride was now due (Completed at 11 pm). The ride was none as I quickly realized a number of problems like way too small gear ratio , crank length (still kids bike crank), seat as a must have :) & much too high handle bar. Well , I wore a helmet ,and realized that it won't work w/o , often colliding twith the frame parts overhead

    

    

    

    

    Fig. 19 : Completion lower and upper bearing

20. The crank was replaced with a solid 3-piece unit with 185 mm arms from Profile racing ( some accessible goodies from my flatland BMX bike) . Was happy that I quickly figured a way how to install the crank into a much smaller frame eye of a kids bike . The test ride was done same day (post-midnight , and some dog walkers started to notice me testing some weird vehicle. No comments by then - just repeated stops and waving heads...) However, while the crank length was now fine (significantly imprving ergonmetry too), the 25 T sprocket had to be replaced.

    

    

    Fig. 20 : Installing the 3-piece crank

21. Sunday, I finally installed a 39T, lowered the handle bars , and dialed in the Saddle bracket and final position of the center of gravity of the vehicle. Sunday-Afternoon , third weekend in the project, I am managing to make my first sucessful Figure-Eight ride on the monowheel. This felt absolutely awesome, and the kids were just sharing the total excitement of their dad.

    

    

    

    Fig. 21 : First successful test ride

22. The 4th weeked was challenged by the goal that we all wanted to go biking together. I had to come up with a bike rack somehow as you won't be able to just walk in to a trailer store and ask for a rack to hold an monowheel (or sth. typically not seen on public roads after 1950's). So , I loaned the key components for my DIY rack from my Ultimate MX hauler (a hydraulic platform to transport the motocross bikes on the back of a truck) , and the rest from the lumber dept. at the homedepot. The bike rack ended up carrying the Monowheel as well as 3 more bicyles.

    

    

    

    Fig. 22 : Figuring the proof of concept for hauling the monster bike.

23. The rest of the Saturday me and the kids were busy giving everything a decent look as we planned to stop by at a bike show in San Jose , which we learned about just the night before.

    

    Fig. 23 : Paintjob to Mono & bike rack

24. Having the bike on the rack we started to hit the roead to the Bike Show in San Jose. The Wheel was just huge, and we had to respond to curious questions of people right at traffic lights . The main event at the SJ history part for the rest of the day was very nice . Many custom bikes & we even had a tag, quite some attention to a working MonoWheel crusing around. We were still somehow off-topic , and our vehicle-type was definetly not on the events list classified. Overall , we spent a very nice day at this event. I forgot my camera at home, but I spotted me riding the MonoWheel in sb.'s video @ ca. 2:10.

    

    

    

    Fig. 24 : Shiny Side Up 2016 show up

25. The fifth weekend since the project started . We make a first bike tour on the bike trail from Morgan Hill and San Jose. The big wheel disclosed a few more bugs , but it ran overall very nice. Was not prepared to do properly mount the camera, but a few shots from the rotating wheel & forward view give you some impression .

    

    

    Fig. 25 : Out on bike trail with family

26. That's it for now. The bike is still being modified , and I am trying to get a better rider on it…
27. 2016-08-07 - Update: Videos on the latest rides

28. Fig. 27 : I could not resist to try out the MonoWheel riding towards pacific shoreline while the rest of my team is playing in the sand on the beach somewhere on the way to Monterey , Ca.

29. 2016-08-14 - Update: Videos on the latest rides

30. Fig. 28 : Practicing static start & testing "pitch plate" steering mechanism

31. Fig. 29 : having at least coaster brake for downhill-rides was a reasonable approach by now. However , this clip shows an accidental brake lockup while entering a turn with some momentum.

32. Fig. 30 : improving stability during left & right turns

33. Fig. 31 : getting very comfortable riding my monowheel by now - after finding techniques to quickly getting in motion from stand-still, as well as getting used to the steering. Getting on is combination of moving the center of gravity forward and cranking - a process which needs quite some torque to generate the necessary initial momentum for the wheel to stabilize (~1 mph is what my wheel approx requires). The problem to manage is gerbling (flipping-over during accelleration or decelleration) , but which can be kept under better control by mass-re-distribution front-back (in-wheel-plane). This helps, because it allows to direct the energy flow into the rotational motion of the outer wheel instead into the energetically more favorable simple raise of the potential energy of the hub incl. rider (…because in rotational systems you store more energy as you account for mass-distribution in addition to just translation of some equal mass moving in a linear fashion). The steering of the wheel is essentially based on re-location of the center of gravity out-of-the-wheel-plane, combined with pushing the crank hard, to compensate the easily possible oversteer. While the latter is mainly through centrifugal forces explained , the physics of the steering itself is based on that any out-of-plane center of gravity applies a torque to the wheel, with its vector oriented either forward or backward (depending whether leaning left or right resp. the direction of the torque vector flips 180 deg). Within the mechanics of gyroscopes, this torque vector describes the orientation of the axis of the gyroscopic precision relative to the current orientation axis of the main wheel. Now, since all 3 axis are orthogonal to each other, momentum conservation alone is responsible for a literally instant turn of the wheel around the vertical axis of the wheel as soon as you lean somewhat lout of plane. In theory this steering can be considered as much more efficient to anything relying on friction to the ground to change directon. However, the steering requires you to master the control this bike more like the roll or pitch of a helicopter, while you are still on the ground. This is unusual , and yet very exciting. Here - the “pitch-plate” function of the heli is comprised by nothing but the frame I am grabbing overhead, periodically pushing sideways.