Making a Rolling Cube : Part 8 - Quick Update

[@RishabhMutha]

It's 2019... Will the part 9 of this come out?

[@longercats]

2020 now

[@Sahil-bx7xg]

@@longercats 2021 now

[@longercats]

@@Sahil-bx7xg let's keep this tradition up

[@Sahil-bx7xg]

@@longercats sure

[@ManzyManz]

@@Sahil-bx7xg 2022, keep this going

[@Mcschism]

Hey Kris, I just found your channel and watched all 8 videos of your Rolling Cube project. I just want to say, as a starting mechanical engineer, that it's truely inspiring to be able to see these kinds of projects being designed and built in a step by step way. I really hope you'll be able to finish it eventually and thanks for making these videos!

[@DoctorBocwurst]

I just want to say, that I really enjoyed following your process so far. I know, that you have the right tools and knowledge to finish it, just don't give up!

[@KrisTemmermanNP]

Leons_KZbin thanks man! No not giving up, there are still some things i can try :)

[@gamen3693]

maybe position all the circuitry at the middle of the cube...

[@weasel.3683]

Eduardo Gamuyao The electronics are not the problem, they weight basically nothing

[@frysinclairmobilemechanic7732]

Hey Kris, It looks to me like you've got to make the flywheel out of a heavier material too. Try milling some steel ones, just as thick. You will be able to store a lot more kinetic energy in heavier steel over the aluminium ones. Think about it. It takes more energy to spin the heavier steel thus giving you more energy/power to move the cube. Hope this helps

[@CSamuel54]

+1

[@DustanWebb]

+1

[@gary.richardson]

Better to increase the speed of the motor and braking for efficiency sake.

[@BenWillBarrows]

This video is fantastic at explaining the issues around developing reaction wheels kzbin.info/www/bejne/apzJc3iXidOWoKs In particular, saturation. It's not as simple as (faster + heavier) = better.

[@MosMeister]

It'll have to use more energy to ramp up the flywheel, or have a slower rampup curve... Anyway, it will get slower or burn ;-)

[@kennethsch2943]

Who else is still hoping for the next part? :D

[@singwolf.]

Kenneth Sch lol

[@jonyjohan8958]

me 2019

[@medienmond]

Me, 2019/2020

[@GGGG_3333]

2020 😢

[@Freglin]

2021 :(

[@Simon_Rafferty]

Very nice cube! However, you are missing a trick! You are only using reaction torque, but you can add to that using the procession torque (which acts perpendicular to the reaction torque) of the other one or two motors to assist. Also, although the brakes give a nice sharp stop very easily, you can achieve just as sharp braking by driving the motor in the opposite direction (depending of course on the current capacity of your ESC) - like EMF braking on steroids! This looses the mass of the servos and brake assemblies. Lastly, if you move your drive motors as close to the centre as possible, with extended shafts to the flywheels, they require less torque to lift. You've no doubt wondered if you could make it balance on an edge, corner, controlled falling? This is near impossible with a simple PID control. However, it's not terribly difficult. (though it takes a lot of tuning). When it's standing on an edge, it has a certain amount of potential energy (as you have raised it's mass). If you spin up the motors to have the same total kinetic energy, when you stop them (ignoring efficiency, drag, friction etc) and the kinetic has been converted to potential, it has just enough to reach the position of balance on an edge. At this point, you switch to a traditional PID to keep it there! Same applies for rolling over without overshooting. I've not built a cube - but I have built the odd 'segway' and inverted flywheel pendulum - which requires the same approach. Feel free to mail me if you get stuck & I'll do my best!

[@weasel.3683]

Simon Rafferty thats exactly what I was gonna say. I have also worked on inverted pendulums in the past and plan on building a flywheel based drone. I would also like to help you finish the project.

[@pierredestin8079]

first, hats off. absolutely impressive. I have nothing to suggest for your weight distribution problem, but even if i did i would hold off, for it seems your mind expresses itself eloquently in solving major design/mechanical constraints. hope to see part 9 soon

[@marcialelizalde3203]

No more videos of this project? :(

[@slimknight_]

Not sure if you're still paying attention to this video but I think that this problem is totally solvable: 1. Make the flywheels out of steel (as said previously) 2. Make most of the other inner cube parts out of Delrin (or hard plastic) These would both save a lot of weight and create a lot more force. Hope this helps!

[@timegaming1085]

Its 2020 when is the 9th part coming out?

[@badekar]

Hey, I know its been 6 month since you published this video, but did you find a solution? and will it come a part 9 ?. I am very impressed with your work.

[@BeaulieuCarl]

What happened with that project ?

[@Enderkruemel]

I also waiting for the next Part. I think the Best way is to buy reversibele Escs. (all blheli s or blheli 32 Escs can Do this. Also some old Simonk should work, But i dont know for sure ) I think some higher speed and more rotating mass is needet.

[@TienNguyen-tp4rt]

Hi Kris, you did all : electronics, mechanical design and make , coding , ...perfect man.

[@marcialelizalde3203]

Part 9?:(

[@dav1dsm1th]

Could you ditch the braking assemblies and use the eletric motors to provide the braking?

[@Westsatsop]

I wonder if a stepper motor would allow braking through motor direction reversal, thus eliminating the weight and complication of the brake.

[@kr0n3r]

Excellent progress so far. I assume that you got your inspiration from the 'Cubli' ? (or they got their inspiration from you ). i have read most of the posts here and they all come back to the same point - the brakes. What I noticed from your build videos is that you are applying a braking force from one side only which will deflect the flywheel to some degree before it starts to resist at which point the friction force will become greater and thus the braking force takes over and gets applied. Would it not make more sense to have a caliper style brake? create a u shaped bracket to hold the brake pads in place with springs. Have the servo arm push against a plate that pushes two pins - one to the brake pad on the servo side and another that goes through the U-shaped holder, pushing against a lever on a pivot arm, which pushes against the brake pad on the other side. This would apply pressure from both sides at the same time causing it to brake quicker and with more stored kinetic energy, wouldn't it ?

[@922bibo]

I have another suggestion regarding the braking. I think using a solenoids instead of servos to push the braking pads would give you a much faster and more stable response

[@CaptMorteza]

Excellent videos Kris. Please don't give up.

[@dylankristian9466]

Legend has it, he is still shedding weight from his rolling cube.

[@psps889900]

How's the control part going? Do you need some help to make it balanced?

[@bensthingsthoughts]

Don't give up you are almost there... You already did such an amazing job !! Well done Sir ! And I hope we will have progress news ASAP :)

[@KrisTemmermanNP]

Ben's things & thoughts I'm not giving up, thanks :)

[@StanislavJochman]

Hey Kris, you can try overdriving brushless motors to make more kinetic energy. And i know that motors will generate heat but you need to overdrive them for just few second so i thing it isn't a big problem. I hope this helps. Stanley

[@dharmikbhavani7272]

There is no need of heat sink on esc those heat sinks are provided because in rc planes or drones require continuous operation but i think in your case it do not require continuous operation so by removing heat sink you will reduce some weight. use something smaller then arduino due there are many small micro controllers out there to reduce weight. i had observed that you have use some sort of big connectors in motor terminals and esc so either you can directly solder it to esc or use some light weight connectors like bullet connectors. and in each cube surface 4 diagonals gives supports to center and hold motor shaft you can make it to only 2 diagonal instead of 4 diagonal making X on each side. you can also use 3d printing for making cube to loss some weight. reduce the size of your esp8266 pcb make it double layer so it may reduce some weight. hope it would help you and waiting for watch it working. have a good day.

[@gigicaly]

For your problem you can try make it lighter by replacing some components with lighter materials like carbon fiber or some plastic, but you can also try increase the speeds of the motors and weight of the flywheels. so when it brakes has more mass and/or speed. Also as an experiment try to use max spin of the x,y motors when trying to use the z motor to obtain lift. Remember the weight "loss" of a spinning wheel? Ah if you want to lose more weight try using glue instead of screws whenever possible and also I see you used a lot of AL casings and covers. Were those really needed or could they have been replaced with plastic materials? Good job doing this I want to build one to someday. Good video series will be looking out for more on this :) Hope you will make the plans available when you finish.

[@maximilianhf3896]

really nice series! I enjoy your editing and choice of music very much. I would love to see the next part

[@mattiasfagerlund]

What if you move the weight of the flywheels further out from the center? Shape it like a T on the outer edge to move the weight?

[@rafaelementable]

Would it be possible to use the motors as brakes instead of the servos? There is feedback already, so it should be possible to throttle reverse until the wheel is still.

[@mthhhjchambefs6629]

Have you thought about counter balancing the cube so it rotates evenly in every direction

[@horle]

Nice to see you're still alive. I'm really looking forward to your upcoming videos.

[@KrisTemmermanNP]

horle Thanks man!

[@zapallalla]

That standup looks so awesome! Thanks for the series so far!

[@KrisTemmermanNP]

***** thanks :)

[@SkyBoiBoiz]

what are the dimension of the flywheel and the square casing

[@srdjansavic2223]

I'm not sure, but I think that shifting the weight to the outside of the flywheel should give you more power. If the material is not the problem you could just swap out that add on ring you have on the flywheel with a ring made from heavier material and smaller inner diameter and keep the overall weight the same. That should work, if I'm right. Hope I helped, good luck :)

[@KrisTemmermanNP]

Srdjan Savic There is already a steel ring on the flywheel. but yea, maybe use some more at the outer edge. Thanks!

[@jamierollinson9354]

+Kris Temmerman Would a single weight on the edge , give more Potential kinetic energy transfer?

[@KrisTemmermanNP]

no, I don't think so. And it will put the flywheel out of balance, so its going to start vibrating a lot.

[@jamierollinson9354]

+Kris Temmerman balance was my concern on the suggestion..bigger wheels maybe in a bigger cage and keep the workings in the centre maybe

[@kentvandervelden]

Wow, how did I miss this build series? Very cool!

[@KrisTemmermanNP]

If only he would finish the damn thing! Thanks Kent! :)

[@xMoZoMx]

hey, have you ever had the time to continue this project? if so, does it work now?

[@DaGeek247]

Your motors can go waaay faster than the speeds you were getting, i guarantee it. Try upping the voltage that gets to them.

[@yangzhou713]

can you upload somewhere the CAD designs thank you very much

[@Bark777]

Any luck since last time?

[@hunardongsson7087]

You have more weight than the kinetic energy stored in the flywheel can overcome. If you add more weight to the flyweel, you still have to add more energy to spin up the added mass. I would just spin the flywheel faster. If your motors can't do it, get bigger ones. I would also _not_ spin the flywheels that aren't needed to make the box flip over. Having them spinning means your cube is _gyroscopically stabilized._ That will make it _harder_ for the active flywheel to do its job, as though you were putting more weight in it.

[@alexeytrofym5475]

Hello! Do you sell these cubes? we need 1-2

[@atabey321445]

no happy end? :(

[@uRea1]

Awesomeness is here, do it my friend!

[@zohar963]

probably someone said it sorry if i am presumptuous but the outer cube shell(the x shape) is made of aluminum can you make it instead with plastic? wouldn't it take some weight off?

[@KrisTemmermanNP]

+Ran Zohar Yes, although not that much, the shell is quite light compared to other parts. But its important that the main body is very ridged. Otherwise the breaking force is absorbed by the cube itself. Most plastics are quit squashy. Thanks for your comment!

[@TheMusicLeecher]

+Kris Temmerman You can try carbon fibre! Stiff, yet light.. Just an idea..

[@qli880]

Please we need the part 9

[@chris_1337]

Awesome videos, love your channel! Quick questions: would using some sort of regenerative braking in the motor eliminate the need for the bike brakes altogether? As in, essentially shorting out the motor windings to get it to stop as quickly as possible? Also, there are quadcopter motors which are "sensored" so you can get a readout of the motor position at all times.. You've probably already thought of these things and figured they wouldn't work for one reason or another, but I figured they'd both be options to lose some components (and, hence, weight). Keep up the great work!

[@KrisTemmermanNP]

+WeBDaEMoN33 Thanks! I tried that in the very beginning. but a break like that isn't strong enough to stop the flywheel. I have sensors on the motors. Thanks again!

[@artursimonian1972]

Part 9 pls

[@JohnnyRottenest]

Rounding the corners could help. Also, I don't think it's balanced. As a test, you could attach counter-weights on the sides without a motor/flywheel to see if balancing it helps. If so, you might mount the motors on the opposite wall of the cube and have a long drive shaft for each.

[@gorynych_56]

what is this for?

[@X3ST3RIN]

we never have part 9 :c

[@PedrinbeepOriginal]

Hi Kris! Awesome work. I only want to recomend you one app for control the cube with wifi: Blynk. It´s free and more easy to use. Greetings from Spain.

[@FalcoDeVries]

Nice vids Kris;) Ben je toevallig nederlands/belgisch?

[@KrisTemmermanNP]

Props & Projects jep, Belgie, Antwerpen, hoor je het zo hard :D

[@FalcoDeVries]

***** ;) je engels is prima maar je zal het altijd een beetje blijven horen;) En ik had wat werk van je gezien op een BE site en toen wist ik het zeker! Ik ben een groot fan van je werk Kris,prachtig om te zien en van te leren.

[@wolfvierbergen1417]

@@KrisTemmermanNP een update geven van dit project?

[@922bibo]

How about drilling holes in the flywheel and the holders to reduce the weight, just like in race cars. Or using carbon fiber for some of the parts to get a lighter structure. You can also use motors with more power but the same form factor as yours so you a higher power to weight ratio.the last idea might be crazy but I just had it in mind lol, you can place the electronics and control units inside a structure just like the one used in a gyroscope where you can eliminate their weight if those have a great effect.

[@KrisTemmermanNP]

Bara Abu Shaqra yes, thats what i'm going to do first. Trim off as much material of as possible. and see where I am at that point. Thanks for your comment!

[@demej00]

Drilling holes in the flywheel would reduce its momentum.

[@gary.richardson]

Balsa wood and glue can reach a higher stiffness to weight ratio for less cost than aluminum or carbon fiber. Sandwiched foam composite should do the same. The only other change is your fastening points would need to be distributed over more surface area via plate doubling.

[@borayurt66]

I really enjoyed this series, and I admire the your maker skills... Having said that, did you really gave up on this? That would be such a shame... I hope you didn't.

[@isticap]

Kris Temmerman, I understand it has been quite some time since your last update on the rolling cube, or as some would call it Cubli. I was looking into this project and it will definitely be a testament to my engineering skills, but I'm really interested in building one. I'm most curious about what stage you are at now in the game. So, before I begin this adventure, could you answer a few questions for me... 1. If you were to do this project again, would you do anything differently? 2. In terms of time estimates, how much time do you expect it to take to build something of the sort? 3. Do you have any recommendations on what materials to start with? I have a 3D printer, would that be useful at all for this project? 4. Any other notes you would like to include. -Andrew

[@chrizlax]

Any chance you can stick more voltage through those motors? Should get you a bit more energy in the flywheel. If you have to replace your ESC, its still easier than rebuilding everything else. I have very much enjoyed the rest of your videos so far, best of luck!

[@KrisTemmermanNP]

chrizlax I still have a big margin on the power. But the speed is to high, the breaks start slipping. so the kinetic energy gets released to slow. Its about finding the perfect balance between everything. And I think the first thing I have to do, is make it as light as possible. Thanks!

[@chernobill123]

can you upload somewhere the CAD designs and the Bill of Materials ?

[@KrisTemmermanNP]

chernobill123 Ill upload it when its finished, I'm still making changes. The current design is to heavy anyway. :)

[@zeytelaloi]

You should consider milling/3dprinting and copper-wiring your own motors and possibly make them an integrated part of the design. If anyone could do it it would be you.

[@AltMarc]

I would suggest bigger flywheels and new breaking system: using flat spring strips (with rubber glued on) going all around the flywheel, the servo tightening the metal strip, the rubber could be cut from tire inner tubings and have some fixing point (at each corner) to pull it away from the flywheel.

[@KrisTemmermanNP]

AltMarc Thanks, Ill think about the break. Although it doesn't looks like it, the flywheels are pretty much the maximum size already. because there is an other flywheel on a adjacent side that has to be cleared.

[@AltMarc]

***** if you got enough power in the motor, make a brass flywheel (density: 8.9gccm-7.8gccm) put most weight in the rim, don't be afraid to put more weight in this part ! Put a new and light weight breaking system in the corners perhaps you can get just a little bigger diameter. Also replace some spacer bolts/screws, with nylon, short the copper wire, cut down the 8 corner parts, try to shape the motor holder, smaller pcb, and a lighter battery. My ornithopter and small helicopters (300 class) don't like to be heavy, so weigh shaving is almost a kind of sport. Trust me, and I didn't talk about drilling holes (almost) everywhere.

[@KrisTemmermanNP]

AltMarc Yes. First thing I'm going to do is remove as much material/weight as possible. There is still quite a lot of room for that. If that isn't enough, I probably will have to rethink the breaks . thanks!

[@imod52g]

Bonjour, le rolling Cube fonctionne ?? 6 ans sans réponse ça fait long.

[@Aussie50InspiredDavidZ]

WE NEED AN UPDATE FOR THE CUBE!! THE CUBE!!!

[@silverhead8243]

Kris, Why did you stop this project, keep trying and moving forward.

[@ITpanda]

Would love to see project finished! Almost as much as if you released all your code with all your projects.

[@KrisTemmermanNP]

github.com/neuroprod/ most stuff can be found here :)

[@RamiAwar]

It's 2020, WE WANT MORE!!

[@Spaghetti775]

I don't know anything at all about any of this, so this might be a completely stupid suggestion, but it seems like you're doing a lot of processing "on-board", so to speak. How feasible would it be to shift some of the parts out of the shell? Then just communicate wirelessly with the motors via individual signals? I don't know how advanced the controlling software is, so presumably you still rely on information from the gyros inside the cube itself. Hope the move goes well.

[@KrisTemmermanNP]

CellDevises All the parts are needed. Even if I do calculations off-board. So It wouldn't make a difference in the weight problem. And I like the idea that its a freestanding device :) The move will be something for next year or so. I bought a rundown building that still needs a lot of work. Ill make a video about that later :)

[@DRKING-ox4ek]

COOL!

[@justinpfeilsticker9904]

I want you to finish this

[@raymondpietersen3176]

Make the wheels heavier to create more momentum!

[@TheMetalButcher]

How about instead of adding a second motor and flywheel, add a turning mechanism to rotate the flywheel assembly. A turntable, like on an excavator.

[@technodreams4714]

Perfect!

[@victorisra7459]

quiero hacer ese proyecto ayuda !!!!!!!!!!!!!!!!

[@Moosetraks21]

cant lose hope, be an engineer. Without problems to solve it wouldn't be any fun.

[@KrisTemmermanNP]

yea, true, but fixing it would mean a complete redesign/rebuild. And i didn't feel much for that at the time. Maybe I'll pick it back up someday :) Thanks!

[@gary.richardson]

There is a book called the handbook of composites that I recommend reading before going any further. The book is about 1,100 pages and is readable from layperson to engineer. I spent 2 years reading it over slowly and learned a lot from it even though there were parts of the math I didn't know at the time.

[@marcusaurelius6607]

it's been FIVE YEARS. so much for all that work, huh, in the end you couldn't pull it off, hahah

[@NwoDispatcher]

make the flywheels heavier

[@FelixRisingOriginal]

Move the heavy ESCs and Battery into the center. Downsize the MCU. Downsize the ESC Heatsink - a small fan would be better than those large heatsinks.

[@peterciurea7771]

don't shave weight, increase the flywheel weight

[@Free_Born]

dont lighten cube, make heavier flywheels

[@BenWillBarrows]

So glad I found this video series. It has been on my mind to try to build one of these since I saw the CUBLI kzbin.info/www/bejne/pJCZoWBnf5poZ7s demonstrated. The problem you have would most likely have stopped me too. I think this confirms my suspicion that I would need to have a lot more maths, science and engineering knowledge than I do to be able to attempt this.

[@9b85e5da6d]

Ugh! These videos are like watching a broken record.

[@RishabhMutha]

A broken record!!🎧🎵🎶🎵🎶🎶

[@OutlawAladdin]

bro is this shit ready or not?!

[@TheTaticelo]

essa merda funcionou ou não????????????????????????????????????????????????

[@FredSena]

It's 2020... Will the part 9 of this come out?

[@xMoZoMx]

hey, have you ever had the time to continue this project? if so, does it work now?