How to Design a Cycloidal Disk in Fusion 360

I've spent quite the considerable amount of time trying to figure out how to properly design a cycloidal disk/gear. Hopefully this can be a helpful resource for others like me.
Check out my new MOSFET modules:
www.ebay.com/itm/Power-MOSFET...

[@lowen74s]

Nice and detailed, thank you. Designs like this are a good example of using parameters too to help prevent mistakes or update designs easier.

[@smellycat249]

I just finished this tutorial and got it printing. I’m excited about the possibilities with this gear reduction. Thank you very much for teaching what you have learned.

[@garysimmons1631]

When I heard you say it took you 2 years to arrive at this point it made me feel better because that's the amount of time I estimated just to get to this point myself. Thank you for your hard work.

[@laurentbreyer]

This is genius! Fantastic work! I can’t wait to see more!

[@coolpratheesh]

great job explaining such a complex concept so easily.

[@brittond14]

Now that was a pleasure to watch!

[@mateussousa3639]

That's the best practical tutorial that I've seen. Thank you so much.

[@anandpatidar1566]

Thank you levi for bringing such an easy uproach. This Video was really help and very easy to understand.

[@daleballance4964]

Love your way of doing it and will be using it to design a star tracker. Thank for your help I've been looking all over for advice

[@rubarb0406]

Dale, how is your star tracker coming? Are you using pins or cylinders with tangent (friction) contacts?

[@macaion897]

Thank you very much from Japan for sharing this great job.

[@MatchaMakesThings]

Exactly the video i was looking for great information. Not just showing how to do it once but how to do it in the future for other options.

[@hamnavoeper]

Thank you! I've been looking for something like this and this was perfect!

[@ddegn]

Thank you, thank you, thank you for making this video! I like to build robots myself and I have a 3D printer on order. A cycloidal gear system is one of the things I want to try to build. This video should help a lot.

[@artantme]

Man, thanks for breaking everything down, i'll try this (hopefully) in january and give some feedback and thoughts!

[@mprotec1]

what happened? Did covid got hold of you??? now is the end of june???

[@DEtchells]

Wow, incredibly useful and well-presented, thanks!!

[@weetzinng2217]

Thank you so much for the detail explanation!

[@Grzzs]

Great job! Thank you for sharing your knowledge that you earned! Well done!

[@janibizjak9966]

What I needed!!! very good tutorial!!

[@brianweighill6371]

absolutely fabulous video

[@ananthkulkarni9478]

Thanks sharing such a valuable data.

[@gppl77]

Man, right after yours utube shows a tutorial for same thing but in solidworks. Theirs is twice longer and not understandable at all. Just a video of someone silently designing something for unclear reason. So, great work! Thanks for sharing, it's very useful indeed :)

[@orlandovillat]

Very well explained. Thank you for sharing!

[@PaulGouldRobotics]

Very good explanation.

[@FilterYT]

Thanks for sharing, it is a great resource.

[@cheungpaul9483]

Thank you very much. I've looking to learn how to design this type of gear in fusion 360

[@pezgallo2406]

Great job and presented well...

[@ivangutowski]

the 'point' tool is in the sketch>create menu. I also found I had to draw the lines and dimensions in the exact order and start/end point as he did, or my model did not react. Got there in the end, thank you

[@gafil001]

I love what you did, thanks. By the way I tryed a different way plotting the equation whitout succès. I'll keep trying.

[@YamShargil]

Great explanation! Thanks :)

[@andyrodrigoalvarado118]

is true, it's not perfect, but it is extremely useful, congratulations dude, thanks a lot

[@dekutree64]

I think this equation will produce the same curve without having to go through the process of setting up the circles and plotting points one by one: x = R * cos(t) + e * cos(N * t) y = R * sin(t) + e * sin(N * t) R = ring radius N = number of pins e = eccentricity t = 0...2pi Then offset by pin radius, same as in the video around 9:00.

[@ScootymcpuffSr]

Thanks for this comment saved me the trouble of having to figure anything out for myself ;)

[@raymondcandiotes4639]

Could you define what t, e, N are please. Other than that looks so more simple. Could perhaps then write a script to draw the profile.

[@noahboursier]

I ended up using: x(t) = (bR + sR)*cos(t) - sR*cos((sR + bR)*(t/sR)) y(t) = (bR + sR)*sin(t) - sR*sin((sR + bR)*(t/sR)) z(t) = 0 where: bR = radius of larger circle on which the smaller spins sR = radius of smaller circle (pin radius) t = 0 to 2*PI

[@bitp1mp]

Is there a way to enter this equation into Fusion 360 to have it draw the spline to extrude?

[@tecks7044]

@@bitp1mp There is an add-in for equation driven curves

[@drewh0208]

Awesome! Thanks! Yea funny how it can take so long to figure something simple out, not just in engineering.

[@magokeanu]

Very nice tutorial, thanks a lot !!

[@TheMrGreenie]

Thanks. This is great!

[@swannschilling474]

Great tutorial! Thanks a lot!!

[@-makina8977]

Thanks ur video really helped me alot

[@thebeginnerelectronicattac8320]

Great video.

[@smoovecee]

Very nice! Thank you.

[@mmawad100]

Real man THANKS

[@BrianBoniMakes]

I just followed your method and it worked. On the points that looked too far apart I just added a few more points between the ones that were there to make it look smoother. It was nice to find this tutorial but all the deleting of constrains is a recipe for errors. I want to experiment making more cycloidal gearboxes in Fusion 360 but there must be a better way, there should be some way to do this automatically like the parametric function in Solidworks where you just plug in the trigonometric function and it plots the line. Keep up the interesting work, I subscribed!

[@LeviJanssen]

I don’t think fusion 360 has any of those fancy functions, I’ve looked pretty thoroughly, but I guess it’s possible. I guess I’ve grown used to the tedium of this design method. it’s great to hear that someone found this useful!

[@BrianBoniMakes]

@@LeviJanssen Got it. You can add a SCRIPT that will plot the points, here's a demo and it's already pretty close to what we need. capolight.wordpress.com/2018/07/02/how-to-sketch-equation-curves-in-fusion-360/

[@BrianBoniMakes]

Here's the math to draw the gear. community.ptc.com/t5/Part-Modeling/wavy-circle-by-equation/m-p/563286#M30034

[@muggzzzzz]

@@LeviJanssen There is a cool software, its called Gearotic Motion. It can generate several types of gears, including cycloidal gears. All you need is just to choose some main parameters and gear ratios. You can find demonstration videos of it on youtube.

[@witheredways4198]

myyy guy , thanks so much

[@nilton61]

Thank you

[@lambsauce5445]

to be clear, offset is the rightmost button in the *modify* menu. when you press O, you'll see it become blue ;)

[@Posiviata93]

Hey Levi, how well is the x360 handling fusion? It seems to do pretty well based on this video.

[@i-make-robots]

Any tips on designing teeth for a strain wave/harmonic gear?

[@Benutzername0000]

I don't understand how at 9:26 you get additional offset half circles, that just don't appear when i do the offset. Care to explain?

[@chriss1914]

i had to do the offset twice, once neg the pos and combine the two

[@Richard-rk1ru]

Would the little edge that appeared at the transition point between the semicircle and the curve after doing the negative offset disappear if you'd have traced the points in smaller increments?

[@adeelsultan9618]

you legend you

[@Zarundo]

If we have a sturdy bearing at the center, is there a point in having the out "teeth" ob the cycloidal disk touch the pins?

[@leotheperegrine]

I saw videos where there is a cycloid on the outside instead of pins, and where the output pins are kept in place. How do you design the outer cycloid relative to the inner one in that case?

[@kristofpucejdl]

Hello. Great job explaining the procedure! It sure isn't easy to convey something like this without confusion. I am wondering (and mind you it might be a premature question given I didn't go the trouble studying the cycloidal gears very much) concerns that sharp edge where the offset cycloid curve meets the circular arc at each side of the lobe. Doesn't it prevent a smooth action? Isn't it likely that this edge is going to get ground off by the time, increasing the gaps in the fit? I did browse through some images of cycloidal discs and most of them seem to have a smooth curve, though they might be approximations.

[@LeviJanssen]

Some cycloids have these points, some don’t. If you simulate a continuous input rotation and a directly proportional slower rotation on the cycloidal disk, you can see that the point naturally follows the edge of the pin perfectly. No matter what position it’s in, the disk will never try to push through a pin. Additionally, load should be balanced on all of the lobes, so the points shouldn’t take any more strain. It’s quite the fascinating thing.

[@kristofpucejdl]

@@LeviJanssen Right, it definitely looks proper in motion. The thing is, or at least my vague idea is, that in the end it's supposed to transmit torque. So there will be forces on either side of the lobe pushing it against the pin (or vice versa). And there my concern is that even though the theoretical motion is ideal, with the torque it will have the tendency to smooth out that ridge. But it really could be a negligible effect - I was just curious about your insight in this. Did you by any chance see the recent installment by NYC CNC about cycloidal gear? kzbin.info/www/bejne/e5XWZWuCach9oK8 They share the gearbox file, but I have no idea what method they used to design their discs and/or whether they are theoretically correct like yours. They seem to avoid the ridges, though.

[@andrewjackson1511]

Hey I am trying to follow this but when it comes to offsetting the splines, fusion won't let me select more than one spline. What am I doing wrong? great video BTW!

[@tadashi_hamada]

Hello Levi, the content is great on this channel. You got a new subscriber. How would you alter the dimensions for the tolerances of 3d printing? Do you simply use offset on the entire boundry of the cycloidal gear?

[@LeviJanssen]

That’s exactly what I do.

[@tadashi_hamada]

@@LeviJanssen Thanks for the quick respond!

[@nostamine2567]

not all heroes wear a cape

[@akshaypadmanabuni2075]

I am trying to di this for a 20:1 but it isnt working out, the line on the smaller circle doesnt move when the angle is changed

[@frederiklueschen3105]

👍

[@wei1224hf]

how to design the center holes?

[@stephanc7192]

Wow

[@user-tk3wh1ud4r]

Thank you for your tutorial!! 9:29 For my trying, it didn't work well :( It came out a different line from you.

[@lambsauce5445]

I've got the gaps on my cycloids. What do I do????

[@mikefyfe7068]

I can draw and assemble all the part for the cycloidal drive, and get the cycloidal disk to turn right. How do you get the Output rollers and shaft to turn with the cycloidal disk?

[@FeartheLess]

hey, did you ever figure this out? im trying to do the same thing, thank you

[@mikefyfe7068]

@@FeartheLess Yes I did. I used this www.tec-science.com/mechanical-power-transmission/planetary-gear/construction-of-the-cycloidal-disc/ To set the inner pins to turn with the cycloidal disk, set your joints, go to edit feature first angle set at 360, Second set at 36

[@FeartheLess]

@@mikefyfe7068 thank you very much

[@varunahlawat169]

Or just fit a spline on the cycloidal equation using fusion360 API. I've almost finished it just a few more hours of debugging left hehe.

[@manujain6582]

Can someone explain why he multiplied the diameter by 2/3? Thanks!

[@illusivec]

diamter*(2/3)

[@bharathkumar1585]

I just started learning fusion this is my first project. Im not able to select multiple curves for offset, it is showing cannot offset a curve. Please help me guys.

[@andreapregnolato920]

without bearings?

[@shanerussell7335]

Isn’t your plot of points part of an elliptical shape?

[@LeviJanssen]

No. It’s an epicycloid

[@flexo23dl]

Спасибо за видео

[@JohanDegraeveAanscharius]

Great, prefer this way against the addins, which don't give you insight. Of course after making this, I will use the add-in...

[@lpjunction]

When the outer ring degenerates to just 3 pockets and the center ring became a 2 teeth gear. The structure became something like a Liquid-Piston.

[@jarisipilainen3875]

16:02 but still not get it.if all pin need contanct it allways. well there it is. do it. let it it spin and then boolean. it forms it? it will be perfect if it wobble correct ammount

[@akshaypadmanabuni2075]

I got the cycloid right but it crashes into the housing, I don’t think the way to create the housing is correct

[@oeq0616]

umm no output drive ???

[@johnk2113]

Hi, I don't think it's a 10:1 reduction ratio. Because you have 11 pins and therefore each revolution moves the cycloidal plate by one tooth. So you need to rotate the shaft 11 times to get the plate to the starting position.

[@KlockworXMusic]

You're wrong, its 10:1. Each time the disk moves one peg, 1/11th of a turn, the input shaft doesn't make a full rotation, it makes 10/11th of a rotation from the previous peg for that same tooth on the disk. That means every time the disk has moved around all 11 pins, one full revolution, the input shaft has rotated (10/11) * 11pins = 10 times. 10 turns of the input for 1 turn of the output = 10:1

[@MichielvanderMeulen]

Great explanation, but please don't use sketch patterns! Feature or body patterns are much better and easier to control.

[@pplpilot]

I wrote a python script to do this in a few seconds... just pump in the values.

[@BrothernutSquash]

Can we please have access to it?

[@pplpilot]

@@BrothernutSquash no, write your own...

[@flowerchild4499]

Wouldn't it be better to use a proper gear-creation tool? The scientific reason for the shapes of the teeth in gears is that there is no friction. Your design will rub and heat/wear ...

[@BaldurNorddahl]

This is not a gear. You can use bearings for pins to avoid friction.