Just a tip. Many of us have ender 3 and have upgraded the main boards. I like to repurpose these old boards. Even the 8 bit creality Melzi boards are great foundations for this type of projects. I'm also a fan of running GRBL32 on a ESP32 over wifi, I even got Marlin running on an ESP32 with a tiny OLED display.
Hi IV Projects, I am really impressed. How you have tested everything. You have done really well. If I understand correctly, you are using a toothed belt with a pulley to increase the wrap angle. This is certainly also practical when you later have to tension the toothed belt. There are also many things which you can still retrofit. I could imagine that interchangeable attachments would be cool. Maybe you could also connect a camera and try a sort by color. I myself am also new to prototyping, for my mechanical engineering studies. That is why I am impressed by your machine. I would not have been able to do that, at least not yet. Really good and informative video, keep up the good work.
Hi Raffael, Thank you very much! Yes the timing belts can all be tensioned without disassembling the robot if the belts stretch over time. The gripper attachment is easy to swap out, so there are a lot of things I would like to try with this robot. The sort by color idea sounds interesting!
Thank you! I'd like to design a 6 DOF style arm as well at some point. The inverse kinematics of a SCARA is a lot simpler, and I would say it's easier to mechanically design a SCARA around 3D printed parts than a 6 DOF arm. I think it would be pretty difficult to design a 6 DOF arm using 3D printed gearboxes instead of more compact metal planetary gearboxes.
@@ivprojects8143 Hey, would it be possible for you to upload a bill of materials? (I can work most from the solidworks files, but I'm not sure about the rest, like belt lengths)
@@miron__ Sure, I just added a BOM to GitHub. Please let me know if I missed anything important. I also updated the SolidWorks Assembly to fix some little things.
Cool project! But I have a question about inverse kinematics: 7:48 is it really necessary to flip "knee angle" when a head passes over the Y axis? It costs a lot of movement and time. I think, it's better to bend arm in one predefined direction. It would work much faster. Please answer and good luck with other projects!
Thank you! It is necessary to eventually flip the knee angle to have the largest possible work envelope, but it could be more optimized. Rather than flipping every time it crosses the Y-axis, it would be more efficient to only flip when needed to reach the desired position.
You can run several steppers simultaneously using the multipstepper library and implement acceleration/deceleration using Newton's equations of motion.
I’ve recently become fascinated by strain wave actuators. Have you experimented with them at all? It seems like a good fit for your z axis since they are not back-drivable.
I haven't tried strain wave reductions because I think they would be very difficult to 3D print. I have seen a few videos of people doing just that, but I've never seen anyone test them for long periods of time to see if they're reliable.
I've been trying to see how the output shafts as connected to the next segment out. I could see the one next to the metal z axisb but not the others. They seem very ridged. Are they just plastic half cylinders with bolts gripping the shafts?
The segments are connected together using printed shaft couplers with inset M3 nuts to tighten them onto the D-shaft. I think the view of the CAD model at 1:08 shows them the best. It took a few tries to make them rigid enough. Hope that helps clear things up.
Thank you very much! Multiple axes can move at once, but the accelerations aren't as smooth. In the intro clip you can see all the joints moving at once.
Thank you! Since these actuators use timing belts, there is no backlash that I am able to notice or measure. Properly tensioning the timing belts is key here, which is why I focused on that a lot. There is a little bit of inevitable deflection under load in the horizontal direction due to belt stretch. This can't really be measured in degrees though, since it's load dependent. I think it could be roughly expressed as a spring constant with the right testing.
I don't remember exactly, but they varied a lot between parts. Parts like output pulleys need stronger print settings than other parts like the arm segments. I can't give really specific advice without knowing what printer you're using etc.
What material did you print this in? PLA? PETG? Did you have any issues with the motors heating and warping the prints since they're directly connected? Any input is appreciated. Thanks!
@@ivprojects8143 Great to hear. I was a bit worried about that. I know they don’t run too hot, especially in an application where they aren’t running for as long or continuously as say, a 3D printer. I just didn’t want to be wasting my time trying PLA. Excellent work. I look forward to more videos from you.
Hey, Thanks for just an amazing video. I would like to ask if this hand was being operated by remote controller or an app or AI. If its using an app, can you please confirm the name
Yes, all the pulleys are printed. I ran out of the black filament at one point so a few of the parts are gray. They are all printed in PLA, however. The output pulleys like that gray one need to be printed at a very high infill value because they experience high loads where the D-shaft interfaces with them. The compound pulleys can be printed at a lower infill value since they don't experience this load.
hello i love your project . i am an newbie in robotics and was wondering if you could ship me the scara robot for learning purposes. not sure how to build the parts thanks
@@ivprojects8143 At first it was SCARA but the complexity and cost of the Harmonic drive. So for now it is POLAR running Duet Firmware. Nema 23 and Nema 34. It may be Dual arm so print 2 columns at once. Based on Affordable Aluminum Trusses.
Cool project man! I’m also an undergrad MechE student and I did a very similar SCARA robot last year, same general set up with the cnc shield, motor drivers, inverse kinematics, etc. Have you looked into actuating a robot arm with BLDCs? I’m working on a 6DoF robot arm with BLDCs and it’s been fun. A lot more torque and smoother and quieter operation although a bit harder controls since they operate on a closed loop system.
That sounds like a great robot you're working on! I absolutely want to try using BLDC motors in place of the steppers, but I haven't bought any yet due to their cost. I'm guessing you're using O-Drive? A 6 DOF robot is on my list as well, but I'm still trying to decide on the design of the forearm in particular. I'd like to do something like the concentric shaft setup used on some of the Kuka robots. Are the gear reductions on your 6 DOF robot 3D printed?
@@ivprojects8143 Yes I’m using two Odrives to drive 4 BLDCs, then I have a nema 17 stepper and a smaller servo that take care of the other 2 DoF. Did that to save in cost because that way I didn’t need to buy a third odrive and the 4 BLDCs handle pretty much all the load. And I got lucky cause I was able to get a student org at my university to “sponsor” the project and buy me all the parts. And yeah for the reduction I’m using 3D printed pulleys with belts. I’m also using a differential drive which means that two motors driving the differential give me 2 DoF and both motors work together to counteract the moments created by the weight of the arm. That way I only need half as much torque from each individual motor and allowed me to use cheaper ones. If you have a Discord or something I can share pics of the setup.
I zeroed it at first, but I accidentally bumped the setup and forgot to fix it. It doesn't really matter in this case, since it's only measuring repeatability.
Hello, I really congratulate you for you neat and smart design. Especially the V-Rollers and counter weight like in a real elevator is a good idea. Other than these, your design seem to be quite modular and each motors/activators could easily be reached and replaced. Normally I printed and assembled another similar model I found here in RUclips, but the biggest problem I faced was replacing cracking/broken parts, ruined electronics/stepper motors. In your design it seems like it won't be that much hard for me, that is why I am thinking about building your scara robot. Could you please give some clues about the elevator mechanism. How can I find such a steel part to mount the sigma profiles onto which the V-rollers will go up and down? And also a little more about the belt and counter-weight mechanism of yours. I examined your BOM but could not see something related to these. Any help would be appreciated.
Sure, you can use my design for your project as long as you provide credit. I'm a full-time student so I won't have time to walk you through the build, but if you have a few questions along the way I'll do my best to answer them. The SolidWorks assembly file should explain how things go together pretty clearly. Good luck!
Ha, was checking out this video due to interest in SCARA motion systems, and its a cool project, but what really got me was the music - same tune I used in one of my videos ruclips.net/video/btdGNpbdXTo/видео.html small world right! btw. you should attribute the music as its under an attribution license..
![FendiDa Rappa - Clock Dat ft. Shamar Marco [Official Music Video]](http://i.ytimg.com/vi/3NBU8KHdxYY/mqdefault.jpg)
Just a tip. Many of us have ender 3 and have upgraded the main boards. I like to repurpose these old boards. Even the 8 bit creality Melzi boards are great foundations for this type of projects. I'm also a fan of running GRBL32 on a ESP32 over wifi, I even got Marlin running on an ESP32 with a tiny OLED display.
Thanks for sharing the CAD and the code....I enjoyed every second of the video!
Glad you liked it!
Hi IV Projects,
I am really impressed. How you have tested everything. You have done really well. If I understand correctly, you are using a toothed belt with a pulley to increase the wrap angle. This is certainly also practical when you later have to tension the toothed belt.
There are also many things which you can still retrofit. I could imagine that interchangeable attachments would be cool. Maybe you could also connect a camera and try a sort by color.
I myself am also new to prototyping, for my mechanical engineering studies. That is why I am impressed by your machine. I would not have been able to do that, at least not yet.
Really good and informative video, keep up the good work.
Hi Raffael, Thank you very much!
Yes the timing belts can all be tensioned without disassembling the robot if the belts stretch over time.
The gripper attachment is easy to swap out, so there are a lot of things I would like to try with this robot. The sort by color idea sounds interesting!
amazing counterweight beam contruction!
Thank you!!
Great job on the design and build. Thank you for sharing with us.
Thank you very much!
You are way too talented!
Thank you very much!
For the algorithm and your works is amazing, very good 👌
Thank you very much!
Awesome project!
I've been working up to building a 6 DOF style arm but seeing this has made me consider if I should give a SCARA arm a go first.
Thank you! I'd like to design a 6 DOF style arm as well at some point. The inverse kinematics of a SCARA is a lot simpler, and I would say it's easier to mechanically design a SCARA around 3D printed parts than a 6 DOF arm. I think it would be pretty difficult to design a 6 DOF arm using 3D printed gearboxes instead of more compact metal planetary gearboxes.
Awesome project! Thanks for share
Really cool project, thanks for posting the files !
Thank you! I'm happy to share the files. If you decide to build one, just let me know if you run into any issues or if there are missing files.
@@ivprojects8143 Hey, would it be possible for you to upload a bill of materials? (I can work most from the solidworks files, but I'm not sure about the rest, like belt lengths)
@@miron__ Sure, I just added a BOM to GitHub. Please let me know if I missed anything important. I also updated the SolidWorks Assembly to fix some little things.
Where ca i found 3d files please?
@@petermarkovic2011 The files are posted on my GitHub page. The link is in the description.
Very cool.
Distance in mm, weights in pounds 🏴☠️
Very cool project!
Thank you!
Great project! Subscribed
Thank you very much! I'm glad you like it.
Cool project!
But I have a question about inverse kinematics: 7:48 is it really necessary to flip "knee angle" when a head passes over the Y axis? It costs a lot of movement and time. I think, it's better to bend arm in one predefined direction. It would work much faster.
Please answer and good luck with other projects!
Thank you! It is necessary to eventually flip the knee angle to have the largest possible work envelope, but it could be more optimized. Rather than flipping every time it crosses the Y-axis, it would be more efficient to only flip when needed to reach the desired position.
Instant Subscription !
Thank you!
You can run several steppers simultaneously using the multipstepper library and implement acceleration/deceleration using Newton's equations of motion.
Good idea, thank you. I'll give that a try.
Great video thank you for sharing
Thank you!
I’ve recently become fascinated by strain wave actuators. Have you experimented with them at all? It seems like a good fit for your z axis since they are not back-drivable.
I haven't tried strain wave reductions because I think they would be very difficult to 3D print. I have seen a few videos of people doing just that, but I've never seen anyone test them for long periods of time to see if they're reliable.
I've been trying to see how the output shafts as connected to the next segment out. I could see the one next to the metal z axisb but not the others.
They seem very ridged. Are they just plastic half cylinders with bolts gripping the shafts?
The segments are connected together using printed shaft couplers with inset M3 nuts to tighten them onto the D-shaft. I think the view of the CAD model at 1:08 shows them the best. It took a few tries to make them rigid enough. Hope that helps clear things up.
This is really well done. I hope you can get multiple axes working at once, just for the cool factor 😎
Thank you very much! Multiple axes can move at once, but the accelerations aren't as smooth. In the intro clip you can see all the joints moving at once.
Very good solid base. Consider programming in a multi-thread environment. It helps a fuck-load when using steppers in combination with anything.
Thank you! I will look into that.
Very big nice! What is the backlash in degrees?
Thank you! Since these actuators use timing belts, there is no backlash that I am able to notice or measure. Properly tensioning the timing belts is key here, which is why I focused on that a lot.
There is a little bit of inevitable deflection under load in the horizontal direction due to belt stretch. This can't really be measured in degrees though, since it's load dependent. I think it could be roughly expressed as a spring constant with the right testing.
good work! where did you use Arduino i did not saw it
what are the settings used for printing?
I don't remember exactly, but they varied a lot between parts. Parts like output pulleys need stronger print settings than other parts like the arm segments. I can't give really specific advice without knowing what printer you're using etc.
@@ivprojects8143 thanks for the reply. I'm using an ender 3 for 3d printing
What material did you print this in? PLA? PETG? Did you have any issues with the motors heating and warping the prints since they're directly connected? Any input is appreciated. Thanks!
It's PLA. As long as the motor current is set reasonably heat shouldn't be an issue.
@@ivprojects8143 Great to hear. I was a bit worried about that. I know they don’t run too hot, especially in an application where they aren’t running for as long or continuously as say, a 3D printer. I just didn’t want to be wasting my time trying PLA. Excellent work. I look forward to more videos from you.
Hey, Thanks for just an amazing video. I would like to ask if this hand was being operated by remote controller or an app or AI. If its using an app, can you please confirm the name
How to inverse kinematic. I use program grblguru to control scara
I'm using a Python program for the IK like I mentioned in the video. I definitely would like to try grblgru at some point, but I haven't had a chance.
It is not SCARA robot but more like GMF Robotics M-400.
I don't think SCARA is a very precisely defined term. Look up the SCARA robots made by Campetella Robotics. They are a very similar style.
code is not foundable many code ... I need invers kinematic code..pls
So many good engeneer solutions in one single project! Congrats :) subscribed with one video, looking forward to see the others :)
Thank you very much!
Amazing job. Thank you also for sharing the knowledge
Thank you very much!
Hanzhen harmonic drive gear , over 30 years experience , strain wave reducer, robot joint
I saw multiple SCARA arms using a (ball)screw for Z translation while you use a belt, did you try both and selected that one ?
Creative video,like it,thanks for sharing :)
Thank you very much!
Did you print the gray big gear at 1:52? (The black one Looks printed).
Yes, all the pulleys are printed. I ran out of the black filament at one point so a few of the parts are gray. They are all printed in PLA, however. The output pulleys like that gray one need to be printed at a very high infill value because they experience high loads where the D-shaft interfaces with them. The compound pulleys can be printed at a lower infill value since they don't experience this load.
hello i love your project . i am an newbie in robotics and was wondering if you could ship me the scara robot for learning purposes. not sure how to build the parts thanks
Is the serial comm code in your github? Didn't jump out at first glance.
Impresive !!! thx alot for sharing
Thank you! Glad you liked it.
Nice
Thanks!
Awesome project. The music is a bit too loud tho
Thanks for the feedback! I'll make it a little quieter next time.
Great stuff! I'm working on a 8 foot tall 6 foot reach cement printer. You gave me some ideas to try. Thanks!
Interesting! Are you making it a SCARA style design?
@@ivprojects8143 At first it was SCARA but the complexity and cost of the Harmonic drive. So for now it is POLAR running Duet Firmware. Nema 23 and Nema 34. It may be Dual arm so print 2 columns at once. Based on Affordable Aluminum Trusses.
I love the music lmfao. It sounds like something out of Portal
Cool project man! I’m also an undergrad MechE student and I did a very similar SCARA robot last year, same general set up with the cnc shield, motor drivers, inverse kinematics, etc. Have you looked into actuating a robot arm with BLDCs? I’m working on a 6DoF robot arm with BLDCs and it’s been fun. A lot more torque and smoother and quieter operation although a bit harder controls since they operate on a closed loop system.
That sounds like a great robot you're working on! I absolutely want to try using BLDC motors in place of the steppers, but I haven't bought any yet due to their cost. I'm guessing you're using O-Drive?
A 6 DOF robot is on my list as well, but I'm still trying to decide on the design of the forearm in particular. I'd like to do something like the concentric shaft setup used on some of the Kuka robots. Are the gear reductions on your 6 DOF robot 3D printed?
@@ivprojects8143 Yes I’m using two Odrives to drive 4 BLDCs, then I have a nema 17 stepper and a smaller servo that take care of the other 2 DoF. Did that to save in cost because that way I didn’t need to buy a third odrive and the 4 BLDCs handle pretty much all the load. And I got lucky cause I was able to get a student org at my university to “sponsor” the project and buy me all the parts.
And yeah for the reduction I’m using 3D printed pulleys with belts. I’m also using a differential drive which means that two motors driving the differential give me 2 DoF and both motors work together to counteract the moments created by the weight of the arm. That way I only need half as much torque from each individual motor and allowed me to use cheaper ones.
If you have a Discord or something I can share pics of the setup.
@@dossantos9498 That sounds like an interesting design! I'd love to see pictures of it.
why didn't you zero your dial indicator?
I zeroed it at first, but I accidentally bumped the setup and forgot to fix it. It doesn't really matter in this case, since it's only measuring repeatability.
Hello, I really congratulate you for you neat and smart design. Especially the V-Rollers and counter weight like in a real elevator is a good idea. Other than these, your design seem to be quite modular and each motors/activators could easily be reached and replaced. Normally I printed and assembled another similar model I found here in RUclips, but the biggest problem I faced was replacing cracking/broken parts, ruined electronics/stepper motors. In your design it seems like it won't be that much hard for me, that is why I am thinking about building your scara robot. Could you please give some clues about the elevator mechanism. How can I find such a steel part to mount the sigma profiles onto which the V-rollers will go up and down? And also a little more about the belt and counter-weight mechanism of yours. I examined your BOM but could not see something related to these. Any help would be appreciated.
Simply exceptional work through out!! thank you for sharing this. I'm subscribed now and very excited to see what ever you do next.
Thank you very much! I think the next video will be a high-speed pen plotter I'm working on.
@@ivprojects8143 Excellent! I am interested to how you address those challenges.
scara sounds like a good idea but in practice the precision of these things is terrible
Can you give me the link of the catalogue you used to decide the dimensions of the steel channel and the rest of the parts?
Very cool, y love this!
Thank you!
I really love the fate music you have in the background
Arduino is so powerful!!!
I agree; it's a great platform for learning electronics prototyping.
Perfect !!!!!
Everything to the down Perfect!!!!
Love that music brings back gooooood memories :) and ur project kicks ass
Please share cad file
Check description.
Dude, thanks can I do this for our engineering project, can u help out?
Sure, you can use my design for your project as long as you provide credit. I'm a full-time student so I won't have time to walk you through the build, but if you have a few questions along the way I'll do my best to answer them. The SolidWorks assembly file should explain how things go together pretty clearly. Good luck!
@@ivprojects8143 I will be more than happy to include You in our credits and references..
Thanks. Is this a project for a class at college?
@@ivprojects8143 yeah 6th sem mini project.. For btech
Great work. I love the "How it's made" kind of soundtruck.
Amazing man! Great video!
Thanks Eddy!
this is awesome
Thank you!
CAN YOU MAKE FOR ME
WHAT IS THE COST OF IT?
Also Marlin is available for scara
No words how good you are !
Thank you!
👍
Thanks!
Do you use forward and inverse kinematics?
The demos in the video are using inverse kinematics only.
Ha, was checking out this video due to interest in SCARA motion systems, and its a cool project, but what really got me was the music - same tune I used in one of my videos ruclips.net/video/btdGNpbdXTo/видео.html
small world right! btw. you should attribute the music as its under an attribution license..
Pls were can I find the full parts
ahora mismo estoy copiando todo para hacerlo
Hi there, great job mate. Do you have an email that i can contact you?