Check out my favorite 3D printers, the K1 Max: amzn.to/3txQUC6 or if you are on a budget, the Ender-3 V3 SE: amzn.to/3FkCPus I hope you enjoyed this video and learned something new! If you'd like to support me making more content like this, please consider supporting me on Patreon: www.patreon.com/howtomechatronics
Thank you for not only providing the design for free, which I am most of the way through printing on a 1mm noz. but also for providing an article that I can read and follow instead of the video. I prefer it.
One of the contributing factors to the low efficiency of the planetary gearbox is the helical gears. Helical gears, due to their nature, cause axial load. Straight cut gears would be more efficient, but less durable because only one tooth is engaged at a time. An interesting compromise would be herringbone/chevron/double helical gears, in which the axial force is the same on both sides of the gear. It still has the advantages of multiple tooth engagement, and quieter operation, but with a neutralized axial load.
Unbelievably impressive. I've been looking up the math on planetary gear boxes and walking away frustrated. This video is in a class of its own. Sub'ed!
The only thing the world forgets is that it owes it all to a Hungarian engineer inventor called József Pigeon. He left Hungary for the World's Fair and stayed in America. He became a collaborator of Henry Ford. One of the Ford T. models received this kind of gearless drive. By gradually braking the outer ring with a lever-operated fairing, the car went faster and faster.
I think the tight running of the gearbox might be due to an inaccuracy in modelling the ring gear. When you do the boolean operation, the addendum of the original gear becomes the ring gear's dedendum, and the original gear's dedendum becomes the ring gear's addendum, and you don't mention in the video correcting for that. You usually have to because standard addendum is 1m, while dedendum is 1.25m. I think I'm also seeing that at 7:04, where the tips of the ring gear teeth stick all the way into the grooves of the planet gears. The grooves of the ring gear look good though so I'm not 100% sure. In my own tutorial I fix this by "press-pulling" the teeth of the original gear to make them longer, and then increasing the ID of the original gear, before performing the boolean operation.
Thank you Antal, your tutorials are great, I gained some knowledge of from them which I implemented here in this tutorial. In your tutorial you explained the thing with the gear's addendum and dedendum really great. I have just checked my model, and I saw that have only added clearance to the ring gear tip (in the normal gear I have added clearance in the groove, which makes tip of the ring gear to have clearance after using the boolean function). So yeah, I'm missing a clearance at the groves at the final ring gear. I always have hard time deciding what information to include and what not when making a tutorial. If there are too much information it might be overwhelming for the majority of the viewers, but on the other hand it's useful for those who want to learn more details about the subject. This tutorial came out to be more on the visual side, to look good and explain the matter as simple as possible. Cheers!
Thank you. This was very easy to follow and explained the subject well. Good job! I appreciate the tests and measurements you did at the end. That's the best kind of science.
It would be awesome to see how you design such complex assemblies in onshape or fusion360. Raw 10 hours video, preferably with comments :). I still have no idea how to design projects where parts interact with each other
Awesome video! I think we’d all love a comparison video of this and a V2 design a bit more optimized for efficiency. I wonder how high you can get and how impactful different design changes can be.
Not sure what torque value for the stepper you used, but the advertised torque of a stepper is the holding torque not running torque, sorry if I missed you actually measuring the running torque of the motor.
Thanks this explanation is one of the clearest available 👍. Unfortunately I do not have 3d printing facility, are any of these finished models available for sale? Thanks
Very intresting, but regarding the efficiency, have you looked at the Nema 17 Torque curve so you know the torque at the rpm you used it at? The motors are usually specified at holding value and then loose a lot of torque with RPM , so you could test directly on shaft to verify what Torque the motor have under the same conditions. Really nice work, I am new to the 3D printed world and it is an amazing place :)
I believe the backlash compensation for the ring gear goes in the wrong direction for the boolean operation. If you are able (or didn't already), try using a negative value in that part.
Really cool video, really helps with understanding the planetary gearbox. Is there a video or the files available for the set in the section of What is a Planetary Gear Set?
I like your cut-out demo gearmotor. I've only seen metal gear planetary gearboxes I've take apart. All had Spur gears. I guess they can't sinter metal helical gears.
I am fairly certain you can sinter helical or other gears, but you probably can’t laser cut anything but spur. Having said that, each type of gear has different mechanical properties, they are not the same.
Wouldn't straight cut gears be more efficient than the helical design you used? If I understand correctly, the straight cut gears are more noisy, but more efficiently transmit torque. It is for this reason that straight cut gear sets are sometimes used in drag racing applications where noise isn't a heavily weighted consideration, but torque is.
Very nice! I'm wondering, if you intended on printing, why not use a double helical design (V-shape) to prevent side forces during load? Currently the side forces will increase the friction between bushing and the washer and reduce efficiency. P.s. your video gives a bit contradicting information: first you say "when 3D printing the parts come out a little bit bigger" (~10:00), later (13:05) you say you change settings because else "the prints are smaller than the model". Seems like more calibration is required there.
Thank you! Well, I didn't want to use double helical design because not they are not that often used in real applications. Yeah, it would be easy to print them with a 3D printer, but they are very hard to be manufactured in reality. Also assembling them can be hard. As for the prints being bigger and smaller, that's a nice catch, it sounds contradicting. The thing is that the prints outer walls are always bigger, but in case of a hole, having a bigger outer wall would results having a smaller hole. Yeah, I could have said that I little better and clearer. Cheers!
Very nice video! thanks a lot. Did you use the rated motor torque or did you measure the real torque output of the motor, when doing the efficiency calculation?
Thank you! I used the rated motor torque which is 28Ncm, but in my previous video, Harmonic vs Cycloidal Drive, I did a real torque test which showed a result of around 20Ncm, you can see that here: ruclips.net/video/IXmCze1GsGU/видео.html
I am from a robotics team currently in FIRST Tech Challenge community and for this season we want to make a planetary gear box to use for two different mechanisms. The torque part of the planetary gearbox for hangning our robot on a truss and the speed part of the system for our intake because we need speed for collecting the game elements. I want to know if it is possible to make this planetary gearbox with one DC motor and a servo or something like this? Can you please help us?
How can the ring gear teeth get correct geometry if they are constructed by subtraction from a regular gear? Does the tool not support a ring gear with correct tooth geometry?
came here to say this, the subtraction method will generate a similar profile but not an optimal one, i suspect a lot of performance issues relate to this design.
In a internal gear hub on a bicycle how is the gear changed? Is it the combination of which part of the gearbox is stationary or how does it work? Thanks
You should sell them assembled, l would buy it directly and would put it on display in my home office. Would love to look at that for ages. It especially fascinates me that when the small white gearwheels are airborne, they perfectly and smoothly enter the other side.
Hallo. Congratulations for a great work and also an excelant job from video point of view as well. I'd like to coment torque efficiency. I think there is no need for disappointment. Torque theoretically should be higher but, we shuld consider that revolutions transition goes from sun to planetary to ring in one planetary box and there are two boxes in a row. So there are 4 stages if I'm correct. At 50% overall efficency that would mean 84% efficiency at one stage which is not bad at all from the point of view that gears are printed, that there is not much of a lubrication and the bearings are simple. In the end again. An excelant job. I'm looking forward of your new projects.
Hello, I was wondering is it possible to add a flow simulation into Visualise? Iv been trying to figure out if this is possible and I figured if anyone would know if this is possible it would be you 🙏🙏
How about converting the weight of your car to pump water to a high tank as a way to store energy?!! 3,000 pounds of you car plus the weight of each passenger can pump a lot of water if we engineer it. The among of energy of that go wasted adding up each day if we don't use it.
Check out my favorite 3D printers, the K1 Max: amzn.to/3txQUC6 or if you are on a budget, the Ender-3 V3 SE: amzn.to/3FkCPus
I hope you enjoyed this video and learned something new! If you'd like to support me making more content like this, please consider supporting me on Patreon: www.patreon.com/howtomechatronics
Thank you for not only providing the design for free, which I am most of the way through printing on a 1mm noz. but also for providing an article that I can read and follow instead of the video. I prefer it.
One of the contributing factors to the low efficiency of the planetary gearbox is the helical gears. Helical gears, due to their nature, cause axial load. Straight cut gears would be more efficient, but less durable because only one tooth is engaged at a time. An interesting compromise would be herringbone/chevron/double helical gears, in which the axial force is the same on both sides of the gear. It still has the advantages of multiple tooth engagement, and quieter operation, but with a neutralized axial load.
Very good coverage of a well-known design. Done better than most other channels. I bet this is an interim video before a bigger project is revealed! 😉
Glad you liked it!
You go it. I first explore the basics than go for something bigger. 😀 Of course, it takes time but I will get there.
Cheers!
Unbelievably impressive. I've been looking up the math on planetary gear boxes and walking away frustrated. This video is in a class of its own. Sub'ed!
Glad you enjoyed it!
The only thing the world forgets is that it owes it all to a Hungarian engineer inventor called József Pigeon. He left Hungary for the World's Fair and stayed in America. He became a collaborator of Henry Ford. One of the Ford T. models received this kind of gearless drive. By gradually braking the outer ring with a lever-operated fairing, the car went faster and faster.
Can you tell me where to read about the details of this Model T gearbox?
I think the tight running of the gearbox might be due to an inaccuracy in modelling the ring gear. When you do the boolean operation, the addendum of the original gear becomes the ring gear's dedendum, and the original gear's dedendum becomes the ring gear's addendum, and you don't mention in the video correcting for that. You usually have to because standard addendum is 1m, while dedendum is 1.25m. I think I'm also seeing that at 7:04, where the tips of the ring gear teeth stick all the way into the grooves of the planet gears. The grooves of the ring gear look good though so I'm not 100% sure. In my own tutorial I fix this by "press-pulling" the teeth of the original gear to make them longer, and then increasing the ID of the original gear, before performing the boolean operation.
Thank you Antal, your tutorials are great, I gained some knowledge of from them which I implemented here in this tutorial.
In your tutorial you explained the thing with the gear's addendum and dedendum really great. I have just checked my model, and I saw that have only added clearance to the ring gear tip (in the normal gear I have added clearance in the groove, which makes tip of the ring gear to have clearance after using the boolean function). So yeah, I'm missing a clearance at the groves at the final ring gear.
I always have hard time deciding what information to include and what not when making a tutorial. If there are too much information it might be overwhelming for the majority of the viewers, but on the other hand it's useful for those who want to learn more details about the subject. This tutorial came out to be more on the visual side, to look good and explain the matter as simple as possible.
Cheers!
@@HowToMechatronics Yeah the balance between brevity and completeness is very tough. I think you did a great job of that in this video, keep it up!
@@antalz Thank you!
You could stick 3 of these together and use it as an astrotracker for star photography once you polar align it.
Thank you. This was very easy to follow and explained the subject well. Good job! I appreciate the tests and measurements you did at the end. That's the best kind of science.
Thank you!
This is the way to do it. Model your parts the size they should be and include the manufacturing tolerance in the slicer.
Impressive gearbox there these are in everything they work so well
It would be awesome to see how you design such complex assemblies in onshape or fusion360. Raw 10 hours video, preferably with comments :). I still have no idea how to design projects where parts interact with each other
Yeah, that's a good suggestion, I will consider it. Cheers!
Me too! Multiple parts and subassemblies are very confusing to me.
Yes please! I say do it in both but if you can’t I would vote for Fusion as that is what I am trying to learn.
Awesome video! I think we’d all love a comparison video of this and a V2 design a bit more optimized for efficiency. I wonder how high you can get and how impactful different design changes can be.
Thanks! Yeah, there will be V2 definitely and I plan to make a comparison video with a 3D printed cycloidal drive. So stay tuned! :)
Very impressive and nice work
Not sure what torque value for the stepper you used, but the advertised torque of a stepper is the holding torque not running torque, sorry if I missed you actually measuring the running torque of the motor.
Very nice design. Thank you.
I've been trying to learn this for a year. Your video is the first on RUclips that's made sense. Thank you so much for including the equations
great video! a comparison of harmonic vs cycloidal vs planetary gearbox ,all 3d printed would be cool!
Thanks, yeah, that would be cool!
This design is very good, very good
Thank you! Cheers!
Such a great video! My next 3D printed project!
Thanks, have fun!
This was super helpful in a sea of planetary gearboxes
Precision is impressive , is it advisable to use it as a waist joint for a robot arm?
Nice work! Aesthetically pleasing as well.
Thank you! Cheers!
Thanks this explanation is one of the clearest available 👍. Unfortunately I do not have 3d printing facility, are any of these finished models available for sale? Thanks
Very intresting, but regarding the efficiency, have you looked at the Nema 17 Torque curve so you know the torque at the rpm you used it at? The motors are usually specified at holding value and then loose a lot of torque with RPM , so you could test directly on shaft to verify what Torque the motor have under the same conditions. Really nice work, I am new to the 3D printed world and it is an amazing place :)
Big thanks for the file !
I believe the backlash compensation for the ring gear goes in the wrong direction for the boolean operation. If you are able (or didn't already), try using a negative value in that part.
Yes, that's right. Nice catch!
Well explained. If you could now apply this method to explaining a an automatic transmission that would be awesome.
Very nice video! thanks a lot.
Thank you too!
Really cool video, really helps with understanding the planetary gearbox. Is there a video or the files available for the set in the section of What is a Planetary Gear Set?
Your content is Awesome!!, Please keep going
Thanks! That's the plan! :)
Another top-notch video. Great work.
sorry, isn't better to use servomotor instead of a stepper motor?
interesting. good work. keep it up.
Thanks, will do!
In automotive applications the ring gear is most often the output.
I like your cut-out demo gearmotor.
I've only seen metal gear planetary gearboxes I've take apart.
All had Spur gears. I guess they can't sinter metal helical gears.
Glad you like it! Cheers! :)
I am fairly certain you can sinter helical or other gears, but you probably can’t laser cut anything but spur. Having said that, each type of gear has different mechanical properties, they are not the same.
So valuable video, i especially appreciate mentioning those rules that designer must follow in order to have working mech.
A worthy video, indeed.
That was really well put together. Great explanation of the design and results.
Thank you! Cheers!
Planetary gear is the king-of-gears
Wouldn't straight cut gears be more efficient than the helical design you used? If I understand correctly, the straight cut gears are more noisy, but more efficiently transmit torque. It is for this reason that straight cut gear sets are sometimes used in drag racing applications where noise isn't a heavily weighted consideration, but torque is.
GREAT WORK!!!
Very impressive!
Great , im waiting for thia
Very nice!
I'm wondering, if you intended on printing, why not use a double helical design (V-shape) to prevent side forces during load? Currently the side forces will increase the friction between bushing and the washer and reduce efficiency.
P.s. your video gives a bit contradicting information: first you say "when 3D printing the parts come out a little bit bigger" (~10:00), later (13:05) you say you change settings because else "the prints are smaller than the model". Seems like more calibration is required there.
Thank you!
Well, I didn't want to use double helical design because not they are not that often used in real applications. Yeah, it would be easy to print them with a 3D printer, but they are very hard to be manufactured in reality. Also assembling them can be hard.
As for the prints being bigger and smaller, that's a nice catch, it sounds contradicting. The thing is that the prints outer walls are always bigger, but in case of a hole, having a bigger outer wall would results having a smaller hole. Yeah, I could have said that I little better and clearer.
Cheers!
ติดตามจากประเทศไทย TH ครับ ขอบคุณครับ
verry informative video thanks a lot
Awesome video as usual buddy
Glad you enjoyed
Great tutorial, so I downloaded the STL archive. But where can I find the partslist??
The only RUclips channel I am always eager to see a new video of.
Excellent
Which 3dprinted small gear will break first when it's under big stress?
Nice project! Where are you from?
Hello
Thanks a lot!
How the torque at backlash test determined ? Is there any standard or rule of thumb ?
Thank you
Very nice video! thanks a lot. Did you use the rated motor torque or did you measure the real torque output of the motor, when doing the efficiency calculation?
Thank you! I used the rated motor torque which is 28Ncm, but in my previous video, Harmonic vs Cycloidal Drive, I did a real torque test which showed a result of around 20Ncm, you can see that here: ruclips.net/video/IXmCze1GsGU/видео.html
Brilliant
Please make a detailed video on how to create own Arduino library..
nice work
Thank you! Cheers!
I am from a robotics team currently in FIRST Tech Challenge community and for this season we want to make a planetary gear box to use for two different mechanisms. The torque part of the planetary gearbox for hangning our robot on a truss and the speed part of the system for our intake because we need speed for collecting the game elements.
I want to know if it is possible to make this planetary gearbox with one DC motor and a servo or something like this? Can you please help us?
Flat earthers do not believe in planetary gears...:D
😅
😂😂
Thank you.
Thank you!
The bushings inner diameter is the same as the dowel pin outer diamater (6mm), how did they fit ?
How can the ring gear teeth get correct geometry if they are constructed by subtraction from a regular gear? Does the tool not support a ring gear with correct tooth geometry?
came here to say this, the subtraction method will generate a similar profile but not an optimal one, i suspect a lot of performance issues relate to this design.
Thanks
In a internal gear hub on a bicycle how is the gear changed? Is it the combination of which part of the gearbox is stationary or how does it work? Thanks
You should sell them assembled, l would buy it directly and would put it on display in my home office. Would love to look at that for ages. It especially fascinates me that when the small white gearwheels are airborne, they perfectly and smoothly enter the other side.
That's a good one! Yeah, to be honest it's really cool and satisfying watching it!
Love ur videos ♥
Thank you!
Hallo. Congratulations for a great work and also an excelant job from video point of view as well.
I'd like to coment torque efficiency. I think there is no need for disappointment. Torque theoretically should be higher but, we shuld consider that revolutions transition goes from sun to planetary to ring in one planetary box and there are two boxes in a row. So there are 4 stages if I'm correct. At 50% overall efficency that would mean 84% efficiency at one stage which is not bad at all from the point of view that gears are printed, that there is not much of a lubrication and the bearings are simple.
In the end again. An excelant job. I'm looking forward of your new projects.
Hi, thank you!
Which M3 bolts have you used ???
I,m an inventor
That's great . Thanks
I'm a mechanical engineering student in my final studies internship and want to get a torque of 30N.m to actuate a ball valve : is it reliable ?
Try printing them with nylon at 150% scale. Probably not worth it if you don't have a 3d printer capable of high temp already
After all this gearboxes, I really wish to see them implemented on a 3D printed robot arm! I hope you can make that wish into reality!
What kind of printer you are using ?
It would be interesting to use it for, say 10,000, cycles with a set load and compare its performance after 🙂
Respecatable sir can i use your 5DOF robotic arm project in my final year research and publication?
Sure, go ahead!
I have noticed you don't answer any of the questions supporter post????
Please make a transmitter and receiver using esp8266 please brother make a video
10:09
Funny, my printer prints something like 0.4% smaller. Which I'm happy with for an ender 3 :D
Wave reducer with intermediate rolling elements - better
T
hanks
Hello, I was wondering is it possible to add a flow simulation into Visualise? Iv been trying to figure out if this is possible and I figured if anyone would know if this is possible it would be you 🙏🙏
0:44 Robot! that my coin. Robowar
How about converting the weight of your car to pump water to a high tank as a way to store energy?!! 3,000 pounds of you car plus the weight of each passenger can pump a lot of water if we engineer it. The among of energy of that go wasted adding up each day if we don't use it.
جيد
👏
👍
weight? (cnc and 3d)
first
Confirmed!
لطفاً فارسی
second
Confirmed!
Please watch your mathematics. "Five times slower" is NOT the same as "One fifth of the speed."
I like this channel.
❤
Awesome, great video❤ @arduinomaquinas 🇧🇷😉👍 thank you bro, subscribed, 👏👏👏👏👏