could you someday do a video explaining the design and building process of an actuator like this in detail? how you create the gears, what is your thought process, etc.
This would be a really awesome video. I would be happy to pay if you designed an online course on design fundamentals, 3d modelling, and understanding electronics.
Very nice. My DIY arm has differential joints in the elbow and the wrist. I use the software "gearotic motion" to generate the gears - it can do spur, bevel, spiral and others, and export them as stl. It's not free though. For the wrist I use bevel gears - there's a bit of backlash there. For the elbow I used timing belts, wrapped at 90 degrees to reduce the backlash.
Very nice indeed!. If you end up with any backlash in the differential sungears, I guess you could always spring load the top one to force it to mesh with the other two.
You're awesome all the technology that you have given to me years ago I was doing this with the robotic arm 30 years ago I wish I stuck with it beautiful advancements God bless you I will watch you closely hopefully I'll talk to you
Hi, I haven't checked your bio yet, will though. Without seeing it, I just have to say I hope you teach. Your method / manner and delivery make you a natural teacher and well, we need more of those. I've been researching, learning to code, learning my 3d printer. I haven't begun my first robot build yet but it's coming very soon. The other "learning" I need to undertake is cad design. I've begun some classes / exercises with Tinkercad but I still have a long way to go. Anyway, GREAT STUFF! Your designs and execution are just spot on AWESOME! ;-) THANKS FOR SHARING!
Hi, what you do is very cool. I am from Russia, I write through a translator, there may be mistakes. I’ve been watching the channel for a long time, I also do diy, but with each of your videos I understand how much I still need to learn and how much I need to grow)
I noticed you had the reduction gearing as part of the motor assembly in earlier joint revisions. That seems like it would have had the advantage of keeping more of the mass closer to the base. Having more of the arm's mass further away from it's base would make it harder to start and stop because of increased inertia.
But you'd have to use stronger belts and pulleys, which increases the weight. You want most of the gear reduction as close to the output as possible, since it reduces the backlash from all the components that are between motor and gear.
This is so cool! I really like the 2 servos for 3 axes with the differential setup. I'm in the design phase of my arm. Although this is a great concept, I'd be concerned with the backlash on this one and am leaning more towards a cycloidal gear system without belts or other gears involved. I'm liking the strong motor and low gear ratio systems idea. You're doing some really cool stuff and I'm learning things from you. Thanks a lot for making the great videos!
outstanding work. For the gears I use the solidworks and believe me it is soooo easy to do it in solid. next time give me the parameters and I will model it in 5 min. you can still open it in F360
You should team up with AnninRobotics and help him implement a differential gear setup into his AR3 open source robot. This would be a total game changer!! 👍🏼👍🏼👍🏼
Wow, I have a love/hate relationship with your videos.. Every time I see a new video of yours i'm amazed at your design choices and decide to redesign myself, adding another week or so to my project timeline. I need to just stick to my end project design and complete it, but you and your damn differential joints keep inspiring new changes. Well done, this dual gimbal motor design looks so sexy. Also, i would like to know what types of motor encoders you prefer using in terms of accuracy / longevity. Do you think prefer rotary encoders of magnetic? thanks!
I thank you and support your channel. Your work is perfect and perfect I specialize in mechanics and I understand it well so the field of design i also have a 3D printer You could work for us a course on how to program robots I lack knowledge in terms of how to control robots and what you advise me to learn Or what are the basic programs in robotics
If you go on KHK Gears website, you can download cad files for the gears they make, I've found that if you download them as STEP files you can import them straight into Fusion 360 and edit them.
The design looks good with regard to backlash and ease of assembly but I believe you need to add additional encoders for the actual robot joints - otherwise you have to integrate the velocity of the motors for determining the position of the robot joints. Then, with a lot of different motion cycles the calculated robot joints might drift from the real joint position due to the differential mechanism backlash or the integration errors.
Hello Skyentific, This is an awesome idea for a manipulator. I have a question. if you had installed the motors directly on the input shaft of the planetary gearboxes, you could have reduced a lot of moving parts, reduced friction, maintenance and increased efficiency. I see that the belt-drives add to the output-torque but that can be added to the planetary also by changing gear ratios. You may have thought of it, any reasons not to follow up on this idea??
Thank you for this question. The bottom part (where there are motors) will be close to shoulder. Like this I put heavy motors close to the arm base. So basically with these belts, I keep extremity of the arm lighter.
Just found this channel, I love it. Haven been thinking about reducing the number of motors by using a power take off system, clutch or whatever. But I have dozens of smaller brushless motors from FPV racing quads I can use, also speed controllers, even thee flight controllers gyro would be useful. My arm will me smaller hehe. Looks amazing your work.
Nice work! Where Will you Place all the Odrives? Isn't Better to build an exetern case? Kind regards PS: are you planning to implement CAN comunication or Is enought serial? Thanks.
How are you dealing with backlash in the bevel gears? Also, what do you intend to use as a slip ring that passes through the upper bevel gear to the wrist?
are those axial motors? I've heard that radial should be used for robotic arms because of the wide "pancake" style of the axial motor giving it trouble adjusting speeds fast. This motor seems to work well though.
Amazing work! Thanks for always sharing great work! I have a question about one potential limitation in this design, in this design you will be controlling basically two axes through the same drive which means you cannot move both axes simultaneously, what is your opinion about this limitation? I believe it could be favourable to have full control for each axis independently so that the robot can move through specific trajectory or do some action without needing to stop one axis and move the other and vice versa. I am very interested to hear your opinion! Thanks so much!
Hi Brother, great to see you here! You can achieve the desired trajectory with two motors moving at once, this can be done at the interpolator software level. You can get it by this line of thinking: joint B rotates when you keep the same speed of both motors. When you have a difference in the motor speed, joint C starts to move. So you basicaly control the speed of joint C on the top of the speed of B. Greetings from Poland, stay safe!
@@jakub-kaminski Thanks so much bro for your reply! I am very disappointed about this guy....he just replies to comments praising him which is not professional. Regarding your comment; so given the design I see he can either actuate the first joint (by moving the motors in the same speed and direction) or move actuate the second (by moving the motors with same speed but opposite direction). Technically given the gear configuration of this design, you cannot rotate both motor with different speeds. This is because the middle gear is connected to both gears on the side so the gears in the side must be rotating with the same speed otherwise of there is a speed difference the middle gear will get stuck and one of the gears will eventually break. Please correct me if I am wrong. This is a fancy design but I do not see any added value except for making a show
It's been great watching you go through this awesome project. I have to admit, though, I'm always worried about two things getting out of control: weight, and cost. Do you have any projections on where those two values will end up when you're all done? Thanks for all the videos so far!
Great question. I try to keep the price as low as possible, but this is difficult. I hope that I would be able to negotiate the good price with seller and to buy in bulk. But for the moment the most expensive parts: Motors x7, each around 100USD, total 700USD; ODrive controllers x4, each around 150USD, total 600USD; cross-roller bearing x4, each around 100USD, total 400USD. So total is 1700USD, and I did not counted smaller and cheaper parts. But I hope to be able to keep the price below 2000USD. I think it should be possible to negotiate the price to 1500USD at least.The weight of the entire arm should be less than 9kg. (I aim 7kg or less).
@@Skyentific Yeah, the cost creeps up. How many spools of print filament would go into one of these? They look pretty solid. Personally I (someday) look forward to having two arms working together, so I always multiply by 2, too.
This differential joint is some clever design! Love it! You mentioned RS-232 for control. Do I understand correctly, that it is TTL UART, not a "real" RS-232?
It looks awesome, congratulations. But I think that it is much easier to program cartesian robots instead of robotic arms. The reason is that using G code we can easily determine positions using straight lines. In contrast, it is always complex to setup positions for robotic arms. But if you know an easy way to do that please let me know.
I already did this ones for my 6dof robot. And I know mathematics quite well. So I am not afraid of it. And there is also very easy way of doing this ;).
Skyentific but don’t you think this one is too heavy at 4kg? Axis 1 is going to have to rotate at least 6kg before even a payload, the cable version was nice and light.
very nice and clean work. I have some issues with your design though: one of the biggest problems to solve in robots is the backlash in the joints. Since you used mostly 3D-printed parts, there will be a good amount in backlash, especially in the planetary and bevel gears. Have you tried to measure the backlash yet and any plans on fixing it? In most industrial robots and machine tools you have encoders in the servos itself (mostly for speed control) and also in the joint itself, so you can measure the actual joint angle with as few distortions as possible. Btw. it's not nessecary to use cross roller bearings, you can also use bearing combinations for the same effect. For example I used an axial roller bearing and a thin ring bearing that fits inside it, with enough room for the frame to go around them. That's also how a lot of indexing tables are guided. You could also pair up two thin ring bearings (61700 series for example) and use those.
Backlash is bad. And I do know this. Check out my video with the cable transmission, I have zero backlash there. I also know that the encoder at the output axis would be nice. But I don't want to build another kaka robot for the kaka robot price. That is why I try to find the way of building cheaper, but functional robot arm. Most humans does not have encoder on their arms, the precision of human arm is low. But thanks to the vision system, most humans are able to perform complicated tasks. The Idea is to make simple and cheap robot arm and couple it with computer vision AI, to get a good performance. About the bearings. Of course you can use multiple bearings, or angular contact bearings. But where to buy them? I don't want to go with the SKF super expensive bearings. I want the solution which my viewers can use. And not the solution which is only available for industry and for high price. If you can give me the link to the not expensive thin section bearings, I would be more than happy to use them.
@@Skyentific I don't know how the situation is where you live, but I've been able to get bearings of that kind fairly cheap on ebay. Maybe I can write more about that later.
Ok, I made a screenshot of the bearing solution I created for my own robot arm: drive.google.com/open?id=1Em7vyypSLhT3EYDBU94yv-Z_61Mf4sKu The axial needle bearing is 60mm in diameter on the outside (just for comparison) and can withstand way more force than is needed, the thin ring bearing holds radial forces and axial forces from the other direction and together they also hold torque against the axis of rotation. Of course it helps that I machine my parts out of aluminium instead of printing plastic.
those slews, where the heck did you find them and to what price? :) I found, but one single one of them costs more than my intentioned total of my robot arm project
Hello sir I have made one Robotic Arm but it does not move in a straight line Between two points but goes straight when there are more points in between. I have been using accelstepper and multi stepper to reach all steppers at same time. Any solution to it. It also jerks between points so how to avoid that. I think that's maybe because of backlash in planetary gear box so maybe that can go off when changed with harmonic drive. Thanks in advance sir.
Thank you for sharing what you do, keep up the good work! I've got to ask, is there public or private funding to this channel? You do so much and keep continuously improving. I'm a big fan!
![Megan Thee Stallion - Mamushi (Remix) [feat. TWICE] [Official Audio]](http://i.ytimg.com/vi/5c0_u9PsrsQ/mqdefault.jpg)
I think the design is even more elegant when it seems so obvious in its execution. Simply outstanding.
The fact that it's symmetric is also very elegant in my opinion.
Thank you!
"Wow" is an understatement. Amazing work sir!!
Great comment!
could you someday do a video explaining the design and building process of an actuator like this in detail? how you create the gears, what is your thought process, etc.
This would be a really awesome video. I would be happy to pay if you designed an online course on design fundamentals, 3d modelling, and understanding electronics.
@@suhaylpatel9471 100% agree
Very nice design. Simple and straight forward
Thank you!
Very nice. My DIY arm has differential joints in the elbow and the wrist. I use the software "gearotic motion" to generate the gears - it can do spur, bevel, spiral and others, and export them as stl. It's not free though.
For the wrist I use bevel gears - there's a bit of backlash there. For the elbow I used timing belts, wrapped at 90 degrees to reduce the backlash.
I am always impressed by the elegance of your designs as stated elsewhere here in the comments. Well Done! and VIVA la Revolution!!
As always, thank you for the update!
It's nice to see this project coming together!
Amazing! Looking forward to see a fully functional arm with this setup!
Me too! I hope this will be soon :)
Awesome that you took up my tip to use human proportions! Can't wait for the next update.
Thank you for this tip! :)
you are amazing! Continue your great tutorials and sharing your ideas and struggles too, this really help to understand.
Thank you!
Very nice indeed!. If you end up with any backlash in the differential sungears, I guess you could always spring load the top one to force it to mesh with the other two.
Yes! Exactly! Thank you for comment!
Very clever design
You're awesome all the technology that you have given to me years ago I was doing this with the robotic arm 30 years ago I wish I stuck with it beautiful advancements God bless you I will watch you closely hopefully I'll talk to you
Nice design! I'm looking forward to seeing the final arm project.
Great job thank you for sharing with us
The Robots are coming !!!
5:56 wow, very cool mechanism :D
From a Fanuc programmer this is really cool !! thanks for sharing your work!
You make great videos. It would be cool to make this 2 axis simultaneous. 👌👌👌
Yes. I will do this! :)
Hi, I haven't checked your bio yet, will though. Without seeing it, I just have to say I hope you teach. Your method / manner and delivery make you a natural teacher and well, we need more of those. I've been researching, learning to code, learning my 3d printer. I haven't begun my first robot build yet but it's coming very soon. The other "learning" I need to undertake is cad design. I've begun some classes / exercises with Tinkercad but I still have a long way to go. Anyway, GREAT STUFF! Your designs and execution are just spot on AWESOME! ;-) THANKS FOR SHARING!
Well.. This is just genius.
Hi, what you do is very cool. I am from Russia, I write through a translator, there may be mistakes. I’ve been watching the channel for a long time, I also do diy, but with each of your videos I understand how much I still need to learn and how much I need to grow)
Спасибо за такое решение. Очень интересно продолжение проекта !! )
Wow that’s looking awesome!
Great work, really think your on the right path, thanks for sharing
welcome sir, i am from india and enjoying ur work
I am very impressed! Such a smart design! Thank you for sharing!
Thank you for comment.
What a beautiful machine. Great work!
I noticed you had the reduction gearing as part of the motor assembly in earlier joint revisions. That seems like it would have had the advantage of keeping more of the mass closer to the base. Having more of the arm's mass further away from it's base would make it harder to start and stop because of increased inertia.
Also would reduce backlash
But you'd have to use stronger belts and pulleys, which increases the weight. You want most of the gear reduction as close to the output as possible, since it reduces the backlash from all the components that are between motor and gear.
great work , thank you .
Thank you
This is so cool! I really like the 2 servos for 3 axes with the differential setup. I'm in the design phase of my arm. Although this is a great concept, I'd be concerned with the backlash on this one and am leaning more towards a cycloidal gear system without belts or other gears involved. I'm liking the strong motor and low gear ratio systems idea. You're doing some really cool stuff and I'm learning things from you. Thanks a lot for making the great videos!
Mount a base plate onto it and gopro! You’ll have an awesome video set up if you combine it with ai facetracking
It gives me a lot of ideas. Thank you!
Great design progress! Just subscribed not to miss the future updates - really looking forward to the next axes. :)
outstanding work. For the gears I use the solidworks and believe me it is soooo easy to do it in solid. next time give me the parameters and I will model it in 5 min. you can still open it in F360
ну ёмаё! это божественно! как обычно!
Спасибо. Приятно слышать такой комментарий от специалиста-робостроителя!
Super!!!! More excellent work. Thank you.
Thank you a lot for your comment! You never miss my videos! I really appreciate this!
Can provide the exactly which motor u are using and link if possible please
Wonderful design, truly inspirational! Keep up the fantastic work!
You should team up with AnninRobotics and help him implement a differential gear setup into his AR3 open source robot. This would be a total game changer!! 👍🏼👍🏼👍🏼
sir.............. i m your huge friend.
thanks for this si
Beautiful work!
Great work !
Thank you! And thank you a lot for your generous support!
Please measure backlash and efficiancy in the next vid! can't wait! great design!
Thank you Damian. I will do this, may be not in the next video, but soon. BTW, I really like your robot arm, and the test results are impressive!
Bravissimo!!!It is "difficilissimo"!!!
really really neat design. cheers. i really like it
Thank you! And thank you for your comment!
Exelente trabajo
Hi! I really like what you do! !!! LIKE!
Can’t wait to see it done. Just a little concerned about the heat that will come out of the motors and go to the z board...
Will see. I plan to install two fans to cool down motors.
super cool and neat!
Great Design
Thank you!
Чувак это божественно! идея супер
Love following your project, thanks for sharing
Thank you for the comment!
Wow, I have a love/hate relationship with your videos.. Every time I see a new video of yours i'm amazed at your design choices and decide to redesign myself, adding another week or so to my project timeline. I need to just stick to my end project design and complete it, but you and your damn differential joints keep inspiring new changes. Well done, this dual gimbal motor design looks so sexy.
Also, i would like to know what types of motor encoders you prefer using in terms of accuracy / longevity. Do you think prefer rotary encoders of magnetic? thanks!
Thank you for this nice comment. I personally prefer magnetic ones. For me it is way easier to work with. And they do a good job.
I thank you and support your channel. Your work is perfect and perfect
I specialize in mechanics and I understand it well so the field of design i also have a 3D printer
You could work for us a course on how to program robots
I lack knowledge in terms of how to control robots and what you advise me to learn
Or what are the basic programs in robotics
Your stuff is awesome. Keep it up.
Thank you
Good job, this is Very good
Thank you a lot!
"Two monthes ago!" You're so endearing. ✌✊👍🍄🐛🐛🌈
If you go on KHK Gears website, you can download cad files for the gears they make, I've found that if you download them as STEP files you can import them straight into Fusion 360 and edit them.
Truly fantastic!
Thank you!
Ждем клон KUKA в настольном исполнении с изгибающимися ремнями и коническими подшипниками подпружиненными, чтобы люфты свести к нулю) удачи!
super...thx.
Amazing! Great creative design, and ideas, look forward to future videos, thank you for sharing your work :)
The design looks good with regard to backlash and ease of assembly but I believe you need to add additional encoders for the actual robot joints - otherwise you have to integrate the velocity of the motors for determining the position of the robot joints. Then, with a lot of different motion cycles the calculated robot joints might drift from the real joint position due to the differential mechanism backlash or the integration errors.
Hello Skyentific, This is an awesome idea for a manipulator. I have a question. if you had installed the motors directly on the input shaft of the planetary gearboxes, you could have reduced a lot of moving parts, reduced friction, maintenance and increased efficiency. I see that the belt-drives add to the output-torque but that can be added to the planetary also by changing gear ratios. You may have thought of it, any reasons not to follow up on this idea??
Thank you for this question. The bottom part (where there are motors) will be close to shoulder. Like this I put heavy motors close to the arm base. So basically with these belts, I keep extremity of the arm lighter.
very nice design, why not using NEMA steppers inside?
that is pretty awesome
Thank you!
very cool
Just found this channel, I love it. Haven been thinking about reducing the number of motors by using a power take off system, clutch or whatever. But I have dozens of smaller brushless motors from FPV racing quads I can use, also speed controllers, even thee flight controllers gyro would be useful. My arm will me smaller hehe. Looks amazing your work.
Thank you for such kind comment!
Nice work! Where Will you Place all the Odrives? Isn't Better to build an exetern case? Kind regards
PS: are you planning to implement CAN comunication or Is enought serial? Thanks.
How are you dealing with backlash in the bevel gears? Also, what do you intend to use as a slip ring that passes through the upper bevel gear to the wrist?
are those axial motors? I've heard that radial should be used for robotic arms because of the wide "pancake" style of the axial motor giving it trouble adjusting speeds fast. This motor seems to work well though.
Ha-ha, differential! Cool!
Amazing work! Thanks for always sharing great work! I have a question about one potential limitation in this design, in this design you will be controlling basically two axes through the same drive which means you cannot move both axes simultaneously, what is your opinion about this limitation? I believe it could be favourable to have full control for each axis independently so that the robot can move through specific trajectory or do some action without needing to stop one axis and move the other and vice versa. I am very interested to hear your opinion! Thanks so much!
Hi Brother, great to see you here! You can achieve the desired trajectory with two motors moving at once, this can be done at the interpolator software level. You can get it by this line of thinking: joint B rotates when you keep the same speed of both motors. When you have a difference in the motor speed, joint C starts to move. So you basicaly control the speed of joint C on the top of the speed of B. Greetings from Poland, stay safe!
@@jakub-kaminski Thanks so much bro for your reply! I am very disappointed about this guy....he just replies to comments praising him which is not professional.
Regarding your comment; so given the design I see he can either actuate the first joint (by moving the motors in the same speed and direction) or move actuate the second (by moving the motors with same speed but opposite direction). Technically given the gear configuration of this design, you cannot rotate both motor with different speeds. This is because the middle gear is connected to both gears on the side so the gears in the side must be rotating with the same speed otherwise of there is a speed difference the middle gear will get stuck and one of the gears will eventually break. Please correct me if I am wrong. This is a fancy design but I do not see any added value except for making a show
What a cool robot! I am curious about what modeling software you use.Can you tell me?
Nice work!
Thank you!
Hello. Have you ever try to use harmonic drives in your projects?
you are the boss
Hi.thanks
When you print mechanical parts. What is the value infill (20%.. 30%..50%.....)?
How ? special gear
Awesome!
Hey, Exellent video!! How do you get those specific gt2 pulleys? Do you have a parametric cad? an you show us how its done or share it? Thanks!
Thank you! I will make the CAD files available for my Patrons, and later, when the arm is finished it will be open source and available for everyone.
There is an open scad file on thingiverse that you can configure to make all of the tooth profiles in any pulley size
It's been great watching you go through this awesome project. I have to admit, though, I'm always worried about two things getting out of control: weight, and cost. Do you have any projections on where those two values will end up when you're all done? Thanks for all the videos so far!
Great question. I try to keep the price as low as possible, but this is difficult. I hope that I would be able to negotiate the good price with seller and to buy in bulk. But for the moment the most expensive parts: Motors x7, each around 100USD, total 700USD; ODrive controllers x4, each around 150USD, total 600USD; cross-roller bearing x4, each around 100USD, total 400USD. So total is 1700USD, and I did not counted smaller and cheaper parts. But I hope to be able to keep the price below 2000USD. I think it should be possible to negotiate the price to 1500USD at least.The weight of the entire arm should be less than 9kg. (I aim 7kg or less).
@@Skyentific Yeah, the cost creeps up. How many spools of print filament would go into one of these? They look pretty solid. Personally I (someday) look forward to having two arms working together, so I always multiply by 2, too.
Please add safety case for belts area.
This differential joint is some clever design! Love it! You mentioned RS-232 for control. Do I understand correctly, that it is TTL UART, not a "real" RS-232?
Yes. You are right.
Nice work indeed. I'm curious about the encoders you used, Can you post a link? cheers
It looks awesome, congratulations. But I think that it is much easier to program cartesian robots instead of robotic arms. The reason is that using G code we can easily determine positions using straight lines. In contrast, it is always complex to setup positions for robotic arms. But if you know an easy way to do that please let me know.
I already did this ones for my 6dof robot. And I know mathematics quite well. So I am not afraid of it. And there is also very easy way of doing this ;).
@@Skyentific Thanks... so maybe you could teach us? I would happily pay for a course on this, or maybe buy a book.
I do plan to make a video about the easy way. The hard way with mathematics would not be interesting for most of my viewers.
@@Skyentific Maybe you should rethink this. Most people who watch your channel would love that, and it would be a cheap easy video.
Can you attach links to your 3d models.
I would really like to 3d print them and test them.it looks awesome
Ждём второй части)
В процессе :)
Очень познавательно! Ждём продолжения!
What happened to the cable version? I think this is too heavy, just for upper arm.
It is still alive, I just decided to try this one, as it is easy to build.
Skyentific but don’t you think this one is too heavy at 4kg? Axis 1 is going to have to rotate at least 6kg before even a payload, the cable version was nice and light.
very nice and clean work.
I have some issues with your design though: one of the biggest problems to solve in robots is the backlash in the joints. Since you used mostly 3D-printed parts, there will be a good amount in backlash, especially in the planetary and bevel gears. Have you tried to measure the backlash yet and any plans on fixing it? In most industrial robots and machine tools you have encoders in the servos itself (mostly for speed control) and also in the joint itself, so you can measure the actual joint angle with as few distortions as possible.
Btw. it's not nessecary to use cross roller bearings, you can also use bearing combinations for the same effect. For example I used an axial roller bearing and a thin ring bearing that fits inside it, with enough room for the frame to go around them. That's also how a lot of indexing tables are guided. You could also pair up two thin ring bearings (61700 series for example) and use those.
Backlash is bad. And I do know this. Check out my video with the cable transmission, I have zero backlash there. I also know that the encoder at the output axis would be nice. But I don't want to build another kaka robot for the kaka robot price. That is why I try to find the way of building cheaper, but functional robot arm. Most humans does not have encoder on their arms, the precision of human arm is low. But thanks to the vision system, most humans are able to perform complicated tasks. The Idea is to make simple and cheap robot arm and couple it with computer vision AI, to get a good performance.
About the bearings. Of course you can use multiple bearings, or angular contact bearings. But where to buy them? I don't want to go with the SKF super expensive bearings. I want the solution which my viewers can use. And not the solution which is only available for industry and for high price. If you can give me the link to the not expensive thin section bearings, I would be more than happy to use them.
@@Skyentific I don't know how the situation is where you live, but I've been able to get bearings of that kind fairly cheap on ebay. Maybe I can write more about that later.
Ok, I made a screenshot of the bearing solution I created for my own robot arm:
drive.google.com/open?id=1Em7vyypSLhT3EYDBU94yv-Z_61Mf4sKu
The axial needle bearing is 60mm in diameter on the outside (just for comparison) and can withstand way more force than is needed, the thin ring bearing holds radial forces and axial forces from the other direction and together they also hold torque against the axis of rotation.
Of course it helps that I machine my parts out of aluminium instead of printing plastic.
can you make a video on Teensy 3.6? looks like a very powerful controller
Nice! Have you considered a sunless compound planetary? Maybe 16:1?
when i try to calibrate brushless motor with reduction and odrive, no complete the calibration in some times, maybe it could you happen?
those slews, where the heck did you find them and to what price? :) I found, but one single one of them costs more than my intentioned total of my robot arm project
On Alibaba you can fond them for around 100USD.
Good job! :)
Thank you!
Could you please tell me the tooth profile and pitch of the belts you are using ?
Hello sir
I have made one Robotic Arm but it does not move in a straight line Between two points but goes straight when there are more points in between.
I have been using accelstepper and multi stepper to reach all steppers at same time. Any solution to it.
It also jerks between points so how to avoid that. I think that's maybe because of backlash in planetary gear box so maybe that can go off when changed with harmonic drive.
Thanks in advance sir.
Thank you for sharing what you do, keep up the good work! I've got to ask, is there public or private funding to this channel? You do so much and keep continuously improving. I'm a big fan!
This is great stuff! Do you want to make a video about how you print the parts? How many walls and what kind of infill?
Brilliant!!!!!