I didn't hear "back drive" once!? TH-cam videos about cycloidal drive gearboxes aren't complete unless the creator says back drive at least 5 times! 😂 Great video
I do industrial robotics for a living, keep the video flowing. Most people in the industry take the actual robot for granted. I am still there, 10 years in, wowed by the mechanical marvel a serial robot is. I am more on the software side of things. Building tools to program them better/faster. You will certainly have some nice challenges ahead when you get to the control, calibration, motion planning part.
Another interesting variation on the cycloidal drive is the compound cycloidal drive, where there are are again two cycloid gears, but they are attached directly, so without an inner ring. One of the cycloid gears has one lobe less than the other, very similar to how a compound planetary gear works. An advantage is that because it has no inner ring, it can be made very small and with very low part count
I love that type of cycloidal drive! I’ve actually managed to design and 3d print one that has an outer diameter of roughly 2cm and is fully print in place (its output is not useful and the whole thing is just for show) while still packing a 36:1 reduction!
I really enjoy your videos. You asked for improvement ideas, and I would love to see you have a go at machining the disks in aluminum on your cnc. That’s one of the big advantages of cycloidals over other high ratio drives like strain wave, planetary, pinwheel worm drives, etc. They can be easily machined on a 2.5 axis system. Even a hobbiest cnc should give better tolerances in aluminum than a 3d printer in plastic, reducing or eliminating backlash and the stuttering. The load bearing will be far greater, improving the torque. And the stiffness will be much greater, reducing the hysteresis. If your little cnc is up to the task, that would be really cool to see.
My favourite bored mechanical engineering narrator comes back with another extremely interesting topic! Who knows where I'll use it? Is it going to be a 20°K mechanical testing machine? Will it end up doing anything for hydrogen tanks? I certainly don't know anything else other than that cycloidal design is a very potent and interesting design :D
I always love your videos, and im certainly not a medical professional, but your fingers look really clubbed. I could be wrong, but TH-cam cant afford to be losing promising new creators these days!
With such a ratio I do not predict any success, but if you are Polish, 40:1 is doable ;) th-cam.com/video/epeQwq-aYV0/w-d-xo.html I'm looking forward to the next parts, great job!
I think requirements for an industrial robot isn't changed from what our ancestors set ages ago :D Rigidity first. Imagine that all our motors are stopped, firmly, they could be removed and the inputs could be welded. Now imagine that you grab the end of your 1,5 meter long robot arm and yank it. Does it moved more than 1-2mm? An industial robot would be like a stone in this case, rigid as f___, so when I'm planning a robot arm I start with the thinking of where is it gonna move when all inputs are locked. For example 1 meret long robot arm, with 3kg payload plus the mass of the arm itself, we can count for at least 10kg of moving mass, which is about 100Nm when it's extended horizontally (and stationay!) When it moves, and we want to stop it fast horizontally, it can easily doubles that, plus safe margin. So that's the mindset of which all DIY robot creators should think when it comes to sizing their parts. ;)
Flexispot Black Friday Sale! bit.ly/4eFbsdK (US)
I didn't hear "back drive" once!? TH-cam videos about cycloidal drive gearboxes aren't complete unless the creator says back drive at least 5 times! 😂 Great video
Back drive! ;-). But seriously at these reductions they aren’t so easily back driven.
I do industrial robotics for a living, keep the video flowing.
Most people in the industry take the actual robot for granted.
I am still there, 10 years in, wowed by the mechanical marvel a serial robot is.
I am more on the software side of things. Building tools to program them better/faster.
You will certainly have some nice challenges ahead when you get to the control, calibration, motion planning part.
Another interesting variation on the cycloidal drive is the compound cycloidal drive, where there are are again two cycloid gears, but they are attached directly, so without an inner ring. One of the cycloid gears has one lobe less than the other, very similar to how a compound planetary gear works. An advantage is that because it has no inner ring, it can be made very small and with very low part count
I love that type of cycloidal drive! I’ve actually managed to design and 3d print one that has an outer diameter of roughly 2cm and is fully print in place (its output is not useful and the whole thing is just for show) while still packing a 36:1 reduction!
Stay tuned!!
I always wondered how plumbuses were made.
No, the video is not long enough. I WANT MORE POWAAAH!!!
If you have a CNC, then make the deforming parts out of metal.
I’m expecting that to be the subject of, or at least included in, a future video
This guy does not sound like he wants to be here.
I love it when I see something that I hadn't heard of before.
I really enjoy your videos. You asked for improvement ideas, and I would love to see you have a go at machining the disks in aluminum on your cnc. That’s one of the big advantages of cycloidals over other high ratio drives like strain wave, planetary, pinwheel worm drives, etc. They can be easily machined on a 2.5 axis system. Even a hobbiest cnc should give better tolerances in aluminum than a 3d printer in plastic, reducing or eliminating backlash and the stuttering. The load bearing will be far greater, improving the torque. And the stiffness will be much greater, reducing the hysteresis. If your little cnc is up to the task, that would be really cool to see.
My favourite bored mechanical engineering narrator comes back with another extremely interesting topic!
Who knows where I'll use it? Is it going to be a 20°K mechanical testing machine? Will it end up doing anything for hydrogen tanks? I certainly don't know anything else other than that cycloidal design is a very potent and interesting design :D
very cool, should put a regular cycloidal drive with the same ratio next to it to show the improvement
I always love your videos, and im certainly not a medical professional, but your fingers look really clubbed. I could be wrong, but TH-cam cant afford to be losing promising new creators these days!
I was going to make the same comment. Hopefully it's just genetic, and if not he already got it checked out.
Would love to see you release those design files in your upcoming video :)
The modern way to summon me
I would like to see this Cycloidal Drive paired with a small drone motor to power an electric bike
Ah more video!
Nice one gonna check this one
Great work as always
Thank you! Cheers!
I feel like this guy doesnt have his intonation finetuned lmao
NICE!!!!! That's so nifty
Looking forward to your follow up(s). Thanks for sharing, cheers!
Another great video!
The amount of math was just right for most ppl I think!
Glad you think so. I had concerns.
It showed!
I would have taken more, but I think it was plenty for most.
im not a doctor or anything, but you may want to have them check your heart and lungs. It looks like your fingers have clubbing. Thanks for the video.
Love it❤
Amazing
Have a look into howimat reducers
Oh I absolutely love that. Really interesting design. Almost a hybrid between harmonic drive and cycloidal.
love it
switch to carbon fiber ASA filament and that thing will be able to lift the whole gym. top notch video and design skills as always 👏
This channel has less than 13K subs. How'd you get an adbertiser so soon? You must have another channel/product out there somewhere already.
I think the parts could easily be lasercut for better strengt of the materials
Just need a laser cutter.
I fear that the ball bearings have too much play. At least if you want to achieve a high level of precision.
I have one
❤
Professional advice: don’t use plastic gears
The cadence makes this impossible to listen to.
At least it’s not AI! 😢
With such a ratio I do not predict any success, but if you are Polish, 40:1 is doable ;)
th-cam.com/video/epeQwq-aYV0/w-d-xo.html
I'm looking forward to the next parts, great job!
This reminds me of a split-ring epicyclic planetary gearset
th-cam.com/video/4gkbb2CBzQI/w-d-xo.html
I think requirements for an industrial robot isn't changed from what our ancestors set ages ago :D Rigidity first. Imagine that all our motors are stopped, firmly, they could be removed and the inputs could be welded. Now imagine that you grab the end of your 1,5 meter long robot arm and yank it. Does it moved more than 1-2mm? An industial robot would be like a stone in this case, rigid as f___, so when I'm planning a robot arm I start with the thinking of where is it gonna move when all inputs are locked.
For example 1 meret long robot arm, with 3kg payload plus the mass of the arm itself, we can count for at least 10kg of moving mass, which is about 100Nm when it's extended horizontally (and stationay!) When it moves, and we want to stop it fast horizontally, it can easily doubles that, plus safe margin. So that's the mindset of which all DIY robot creators should think when it comes to sizing their parts. ;)
But you gotta also move it somehow! Great points though.