[MOC] 6 Degree of Freedom Robotic Arm

[Glaysche]

Update:  Here is the latest version of the arm:

 

Studio model available for free on Rebrickable: https://rebrickable.com/mocs/MOC-109607/glaysche/6-axis-robotic-arm-mk-2/#details

 

 

Original post starting here:

 

This is my first MOC and first time posting on Eurobricks.  A few months ago, I got back into Technic by buying the Liebherr set (42100).  I decided to design my own 6 degree of freedom robot.  In doing research and experiments for this project, I was inspired by this post I found here on Eurobricks: 

 

 

I ended up making a lot of different choices in my model.  Here is the big picture:

 

I wanted to optimize for three things:

• It should work
• Maximum range of motion -- the wrist axes should be able to turn indefinitely and other axes should have as much range as is possible
• Built as compact as possible

Here's the video with poor production values (and using a simple BuWizz profile to control it):

https://www.flickr.com/photos/188456966@N07/49900167801/in/datetaken/

This was quite challenging to optimize it in this way.  First of all, in order for the wrist joints to spin, they needed to be axle driven -- pneumatics won't work.  In the example of the grabber at the end, I put an axle through the turntable to control the LA for the grabber.  When the grabber turntable is spun, however, the grabber will open or close in an undesired way.  To counter this, I used a differential and the correct gearing to compensate.  Here's a picture of that:

This is difficult to see in there.  There is a new 2 piece differential (65413 / 65414) inside the turntable.  The gears turn the differential at half the speed as the turntable which makes it so the grabber doesn't move when the turntable moves.  The control for the turntable and grabber are axles on both sides that go through the next set of turntables seen here:

The axle turns the perpendicular axle that goes through the turntable and is driven by another axle in the next stage.  The stage has the same problem as the previous one where flexing the wrist will cause the grabber to turn and open or close.  This is compensated with an additional two differentials and another set of gears.  This one is particularly compact.  Here's a picture of that stage:

The lage gear there is a 28 tooth double bevel which is needed to get the correct gear ratio to the 28 tooth differentials.  Pulling the side off of that, we see:

There are three L motors (22169) here to control the grabber and first two degrees of freedom.  I particularly like the gearing in the front of this section for the compensating differentials:

As noted by Hanso in his series of posts, the end of this arm is actually quite heavy.  I would have preferred to send all three functions though the next turntable but I could not find a solution that sturdy or compact enough.  This weight caused trouble for the next turntable.  To make this work, I used two turntables, one in compression and one in tension.

There is a beam that goes through the center of the turntables that takes some of the load here.  The next stage is a another stage that I would like to be able to turn forever so I don't want the wires to get twisted.  In order to accomplish this, I also put the battery box on a turntable that turns with the rest of the arm.

The wires go through this stage and don't get twisted.  The humerus (in LBG) is very small (5x5) and it was a bit challenging to get the gears in there.

The motors (XL 22172) for those next two axes very neatly attach there.  The base has the last two degrees of freedom.

The spin axis here uses a mechanism very similar to the Rough Terrain Crain (42082).  The final axis uses worm gears to drive the turntables.  There's a lot of force on this axis and I experimented with a variety of different gearing.  The thing that worked best is displayed above.  The center piece is driven and it gets geared down symmetrically to the axles that have the worm gears.  Axle support was a bit tricky here:

(Still waiting for the remaining black 7x11 frames to clean up the look here.)

The 2nd battery box fits pretty nicely here:

 

In conclusion, I think this works pretty well.  I haven't explored the software at all yet -- just used BuWizz as a proof of concept.  Many of the motors would work much better configured as servos.  I wish Lego would release an SDK for Control+ so I could write an app similar to the Liebherr controls.  My day job is actually a software developer so it would be pretty fun to build a full featured app to control this.

Edited May 3, 2022 by Glaysche

[Glaysche]

I made many updates over the last few months.  Here is the big picture:

 

I designed new jaws, made a cleaner way of transferring an axle through the turntables at the wrist joint, completely redesigned the arm rotation gear train, made the inclination joint much more robust, added "spring balance" shock absorbers inspired by Akiyuki's robotic arm, redesigned the gearing for the whole arm tilt to no longer have worm gears, switched to the Spike Prime angular motor for rotation, and designed a brand new base that is much more solid.  It's been a lot of fun.

 

Here are a few more pictures:

 

I think this is right at the edge of what Lego Technic can do.  In particular, the bottom tilt joint is a lot of stress on the parts.  In order to minimize this as much as possible, I drive each of the 60t turntables with 2 12t gears which get geared down a few more times so that the XL motor can drive it.  The axles are very well supported in this gear train.  Some of my intermediate designs permanently damaged some axles in operation.  As always, I'm sure there are more improvements to come.

[Glaysche]

I have made a lot of progress on this robotic arm.  At first, I did a bunch of optimizing of various structures, made it look a little better with better colors, and a few other things.  I ended up with this:

This is still basically the same as before, just incrementally improved.  The big problem here is that the final couple parts of the robotic arm are too heavy for the large turntable and the way the construction works is very cumbersome to take apart.  In order to get the top part of the arm apart, it needs to be taken apart into many small pieces -- not very modular.  This was the best I was able to build with 100% pure Lego pieces.  The next step was to use a 3D printed part designed by @efferman that allows transmitting 3 functions through a turntable:

 

This can be found at https://www.shapeways.com/product/BGBRHWES5/three-functions-through-turntable?optionId=138357654.  This was a revolution.  I was able to make the end of the arm much smaller and lighter.  It works much better and is now a very modular build.  Here's what the new one looks like:

The motors have moved to the back of the arm for good balance.  This allowed me to move the hub down onto the arm which reduces stress on the bottom joint.  It also is very modular.  The red pins and parts typically indicate the parts to remove to take it apart into big pieces:

Every joint now works well in my testing.  Software is now a challenge.  The hubs pictured here are the Spike Prime hubs.  I'm actually using them with the Robot Inventor firmware for my tests.  RI does not allow controlling multiple hubs simultaneously which has some work-arounds.  It also has some incompatibilities with the L and XL motors I used -- you can't read the position on those.  I could convert to use just standard technic hubs but the form factor is not nearly as nice.  I'm still figuring out my options here.  I have an email into Lego support about the incompatibility problems with the L and XL motors.  We will see what they say.

 

Retrospective

 

Building this had different challenges than I thought it would when I started.  I thought the big challenge would be designing the gear trains -- and they are complicated.  I use 6 different differentials and precise gear ratios to eliminate coupling of functions sent through turntables.  The most complicated is the top unit:

This gearing allows rotating the turntable without unwanted movement in any of the other axes.  The real challenge was the limitations in the strength and rigidity of the lego parts themselves.  Every time I was able to improve bracing, make it more rigid, or have better form locking, everything worked better.  I went through many iterations coming up with better structures.  There are still things that are not as good as I would like but I think it is pretty good right now.  I also found wire routing challenging.  A lot of work went into routing wires well which you can see in the above picture and a couple others:

That last picture also shows a color sensor I am using for calibration.  I haven't been able to build a full calibration routine because of the software limitations I mentioned earlier. Another thing that has been challenging that I didn't suspect was something I'll call "color engineering".  I want the colors to be clean and look nice.  I don't want to see blue pins or axle pins anywhere.  I ended up re-engineering different parts to eliminate the blue axle pins.  I couldn't eliminate all of them so I did end up buying some of the expensive black axle pins on Bricklink.  When I first built this, almost all parts were common and purchasable on Brick and Pieces but to get the colors right, I needed to buy rare parts.  Some of the lime green parts are hard to find.

 

Anyway, I wouldn't call this project done but I think most of the work now will be software related so if I end up posting about it, I will probably move it to the Mindstorms forum.  I posted this here because the Mindstorms components are pretty much incidental right now.  Most of my development was done with Technic hubs.  It was a super fun project.  I would love to hear any comments and suggestions on improvements.

Edited January 30, 2021 by Glaysche

[iLego]

This is a great project !!

I think I will have to read the topic a few times to fully appreciate all the details.

Building large and yet functional structures out of Lego is always a challenge. Your dedication and persistence is inspiring.

[Glaysche]

1 hour ago, iLego said:

This is a great project !!

I think I will have to read the topic a few times to fully appreciate all the details.

Building large and yet functional structures out of Lego is always a challenge. Your dedication and persistence is inspiring.

Thank you very much!

 

One other thing I thought was interesting:

This is the main vertical arm.  The basic cross section is 5x5 studs.  I needed a motor in it so I was able to fit an L motor inside.  The motor is actually part of the structure providing strength.  There are 4 12t gears driving the 2 60t turntables.  This made the gear train much more robust.  When I only had 1 12t gear per turntable, it sometimes would skip.  The 28t gears were the largest I could fit in that spot which helped make the rest of the gear train lower torque.  I used this same technique for all the high-stress joints.  The highest stress joint on the bottom has the most complicated gear train.  It has 16:20, 12:36, 12:20, 8:28, 12:60 gearing packed in the base to allow the XL motor to drive it.

[Glaysche]

I did some improvements.  Here's the overall picture:

There were 3 big problems I wanted to solve.  First, the connection at the top was relatively weak and relied on friction to hold it together.  The new one is much better and form locked together with a simpler, more robust gear train:

I think it may look a bit better aesthetically, as well.

The second big problem was the tilt joint at the bottom sometimes struggled under the load and was geared down such that it was already pretty slow.  It was powered by a PoweredUp XL motor.  The solution was to put in 2 XL motors for more power and change the gearing to make it faster.  The motor module looks like:

Which goes in this spot:

This is less geared down than before and overall works much better.

The final problem to solve was the rotation axis on the base sometimes skipped gears under high acceleration.  This was fixed by driving the turntable with two gears and building a new gear train and motor mounting to support that.

Because the space that was previously used to hold the hub is now filled with motors, I needed a different way to mount the hub:

As a bonus, this provides a bit more counterweight and helps the arm stay a bit better balanced.

 

As I mentioned in a previous post, this is pushing Lego to its limits.  Here's an axle after much abuse.  This is one of four axles driving the two turntables for the bottom tilt axis -- the highest stress point in the whole robotic arm:

I think this happened as the result of me mis-assembling the joint and having this axle taking the brunt of the stress instead of equally sharing the load with the other 3 axles.  It's a cool looking failure, though.

Comments, criticisms, and suggestions are always welcome.

[Jeroen Ottens]

Wow, such an impressive model. And hats off for the continuous improvements.

[2GodBDGlory]

That is a very impressive, complex model!

I don't know if it would be possible in the context of your model, but replacing that twisted brown 5L with stop with a standard LBG 5L axle may make it stronger. Axles with stops seem, in my experience, to be much weaker than the stopless ones.

[Glaysche]

16 hours ago, Jeroen Ottens said:

Wow, such an impressive model. And hats off for the continuous improvements.

 

1 hour ago, 2GodBDGlory said:

That is a very impressive, complex model!

I don't know if it would be possible in the context of your model, but replacing that twisted brown 5L with stop with a standard LBG 5L axle may make it stronger. Axles with stops seem, in my experience, to be much weaker than the stopless ones.

Thank you for your kind words.

 

I tried the standard LBG 5L axle and it seems to be working well so far.  Thank you for the suggestion.  I was using the axle with a stop because I have always try to use an axle with a stop when it fits.  It eliminates one way the axle can slide.  In this case, I think it will be perfectly ok to use a standard axle.  What would be ideal in this situation would be an axle like one of these except be 5L long -- the solid bit would be in the middle, 2l axle on each side:

This application has (pin hole, 12t gear, pin hole, 28t gear, pin hole).  All the stress happens in the middle segment between the two gears.

[Glaysche]

After replacing the axles w/ stops with standard 5L axles, things worked better but there was still some strange behavior under high stress.  That's when I noticed an issue with how the vertical arm was attached to the base.  Here's a mock-up of that in colors that are easier to photograph:

The arm is attached using both 5x7 L beams (inside the turntable tabs) and the 3x3 T beams.  What I noticed under high stress was something that looked like this:

The structure was held together with friction and started to pull apart under load. To fix it, I made this piece:

And installed it inside:

And added some more reinforcements:

With these changes the bottom tilt joint now moves much smoother.  There is also less creaking during operation.  The lesson I keep learning is that there is never too much overkill when it comes to solid structure.  If you can find a way to get better cross bracing, everything ends up working better, every time.

[Glaysche]

Here is a minor update.  I added some reinforcements and improved the mounting of the top of the "spring balance" shock absorbers.

You can see the triangular bracing in this picture.  It probably makes a small difference.  One improvements I really liked was the triangular bracing for shock absorbers:

I was able to fit a half beam between the bump on the L motor and the 5x7 frame.  The motor locks the beams in place.

[Glaysche]

I did a big update to the bottom tilt axis.  This axis would work well until something stressed the arm like driving it into the end stop.  At that point, it would often skip gears and twist the axle.  I took my own advice of never having too much overkill and fixed it for good, I hope.  The high-stress axle / gear that was breaking was the axle and 12t gear that was driving the turn table.  The previous design had two 12t gears on each turntable for a total of 4 12t gears driving the axis.  I changed that to be 4 12t gears for each turntable for a total of 8 gears driving the axis.  This should reduce the stress on each axle by about a factor of two and hopefully operate with good margins going forward.  This is what it looks like now:

You can see 3 of the 4 28t gears that are attached to the 12t gears driving the turntable.  With this change, I was able to eliminate the "spring balance" which makes me happy because that's the only truly rare part needed to build this model.  There are still some rare lime green parts but if you change the color, this entire model can be built with common parts available on Bricks and Pieces (plus the one 3D printed part that was discussed earlier).  I also added better bracing and generally made things more robust.

Zoom into the base:

And pull it out of the robot.  You can see a couple of the 12t gears on the far turntable.

Take the outer axle support off and you can see 3 24t gears that connect the two sets of gears together.  You can also see the 8t gear at the bottom which is given by internal gears which are driven by the 2 XL motors.  This gearing gave me an extra 3:1 gearing in the system.  I compensated that by changing the internal gears to not gear down as much.

Take that layer of gears off and you see the 4 critical axles on this side (identical on the other side, of course).  I use the 28:8 gears because it reduces the torque as much as possible as close as possible to the load.  The other axles and gears have never had any problems with this design.  The extra 24:8 step down is also very important because the axle that goes into the internal gears is relatively long.  Having high torque there would cause a lot of twisting of the axle during operation which would make the motion less smooth.

This ended up working much better than before.  Here is a 13 second video of it moving: https://www.flickr.com/photos/188456966@N07/51153949023/in/dateposted-public/ (Embedding a video from Flickr didn't seem to work so here's the raw link.)

 

So overall, there are now 38 gears in this gear train including the turntables themselves.  Is it too much overkill?  Naw, probably just the right amount of overkill.  I'm super happy with how well it is working and how robust it has been.

Edited May 2, 2021 by Glaysche

[Glaysche]

I have finally spent the time to make a 3D Studio model of this.  The original model was made completely in real bricks through a trial and error process. I made the .io file and parts list available for free on Rebrickable:

https://rebrickable.com/mocs/MOC-84097/glaysche/six-degree-of-freedom-robotic-arm/#details

Here's a render:

This was my first experience with Studio.  It definitely presented some challenges but it was better than I was imagining given all the negative comments I have read on various forums.  The biggest challenge for me were missing parts or parts with errors in them.  There seemed to be fixes for all of these but it appears it takes many months for those fixes to get into the main release.  My model definitely has some issues.  There are several collisions -- mostly gears not meshed properly but also a couple others I haven't figured out yet.  The 3D printed part from @efferman is not in this model yet.  The part for the color sensor I found didn't render properly so I left it off the model.  I intend to fix these when I figure them out.  I'm missing the wire clips that are only attached to wires. I also didn't model the wires.  I probably won't fix that unless it gets much easier to do in Studio.  I used blue 3L pins and axle pins in the Studio model even though I used black ones in the real model.  I thought that would make it clearer.  If anyone plays with the 3D model, I would love to hear constructive criticism on how I can do a better job.

I learned a couple things with this exercise.  First, the model has over 4000 parts, almost 200 of which are gears.  It uses 6 of the new differentials which were critical to decouple the function going through the turntables from the motion of the turntables. About half the parts used are pins.  I tried to make the construction as solid as I could.  Part of that was using every pin hole available.  It may be considered overkill but every time I made the structure more rigid and solid, it worked better so I think it was good.  Almost every part is functional.  There are less than 200 parts on the robot dedicated solely to decoration (mostly the tiles and half pins they are attached to).  I tried to make it look reasonably good by changing colors of the structural members themselves.  In the process of building the Studio model, I came up with several ideas to improve the structure.  Some are already incorporated in and some will require some testing before putting into the model.

I don't think I will make step by step instructions for this model unless there is strong enough demand.  The parts to build this including 2 Spike Prime hubs and 8 motors are quite expensive so I don't imagine many people will want to build this.  I hope I'm wrong.  I would love it if people took up the challenge.

Edited August 7, 2021 by Glaysche

[Mr Jos]

Well done making it in Studio, certainly for a first model. 

Insane that is has 4000+ parts and still being as strong! At this moment not much time to look more at it, as I'm busy modeling my own next MoC in Studio as well, but as you said, it has its limitations, and don't know if I will succeed this time (pneumatics!~).

You have chosen a very nice color design, wondering what your next project will be.

[Jonas]

Looks both mechanically and aestethicly nice!

I have a question to that special 3D part. Which of its variants offered on the Shapeways page do you use? There is a large variance in their prices, so I would like to know your experience. 

[Glaysche]

1 hour ago, Jonas said:

Looks both mechanically and aestethicly nice!

I have a question to that special 3D part. Which of its variants offered on the Shapeways page do you use? There is a large variance in their prices, so I would like to know your experience. 

I bought the “Black Premium Versatile Plastic”.  That’s the most expensive one.  I wanted a black part and chose that one without any real research on my part.  The part seems quite solid.  It has a matte finish. It worked right out of the box with trimming, sanding, or touch up required.

 

oh, I did update the Studio model on Rebrickable. The V3 model that is up there now has slightly improved structure, fixed collisions, and now includes the 3D printed part.

[Glaysche]

I updated the model on Rebrickable to V4.  I finally figured out how to import the Spike Prime color sensor properly.  I made a render of some of the most complicated gearing in the robotic arm:

This takes the output from 4 motors and drives the turntable and the 3 functions through the turntable.  Using the differentials and exact gear ratios, you can decouple the functions going through the turntable.  If the turntable rotates 90 degrees, the three functions going through the turntable also rotate 90 degrees so the two wrist joints and the grabber don't move when this turntable rotates.

[ninoguba]

Got any videos of the arm in operation?

[Glaysche]

My videos are not very good.  I posted a couple in this thread. I might be able to get a better video this weekend.

[wissamms]

Impressive job!

[Glaysche]

It's been a long time since I've posted on here.  I have still been making progress during this time.  Most of my changes were inspired by new parts released by TLG.  The first major change was inspired by this part on the new BMW motorcycle:

This shock absorber has much higher travel than any of the other Lego shock absorbers so this would make a really good spring balance to improve axis 1 -- the main tilt axis on the bottom.  Previous iterations of this design had used the shocks from the Unimog set.  Those didn't prove to be that effective.  This one was a little better.  Here are the results:

With the new spring balance, I was able to change the gearing on the axis to make it faster.  It also got a bit smoother.  This was the end of the major improvements on this version of the robot.

To see it in its final form with a bunch more detail and freely available Studio file, see https://rebrickable.com/mocs/MOC-84097/glaysche/6-axis-robotic-arm/#details.

 

The next big innovation came with this part:

Upon seeing this part, I immediately knew I would want to use it to simplify and strengthen the "humerus".  With it, I could make the attachment at the bottom much stronger and the whole thing simpler.  Here is a comparison:

The new one is on the right.  This eliminates the spring balance (you'll see why in the next section) and most importantly, the connection at the bottom is 5L wide rather than 7L wide of the old one.  This required me to redesign the shoulder component.

To help with naming, this picture might help:

This next version of the shoulder started as a blank canvas.  I couldn't simply modify the old one to accept the narrower humerus.  There were gears in the way and much of the geometry used frames that are not available in slightly smaller sizes.  I did a lot of experimentation trying to fit the gear train into the smaller size and still be able to fit the gear train for the rotation axis (more on this in a bit).  The real innovation came when I realized I could move the tilt motors to be closer to the tilt turntables.  This ended up making the tilt axis much more powerful, smoother, and simpler.  It was powerful enough that I no longer needed the spring balance which allowed me to simplify things even more.  Here is a comparison of the old and new shoulder units:

The new one is on the right.  The high stress part of the gear train is similar to the old one but instead of going through a pile of gears to get to the two XL motors, I attached two large angular motors as close to the turntables as I could.  This new gear train is much more efficient.  Here is the comparison of functionality:

It works much better without needing the spring balance.

My next challenge was to make the rotation axis (axis 1) work better.  The old rotation axis used 2 12t gears to drive the 60t turntable.  That worked but it felt a little janky at high acceleration.  I wanted to fit 4 12t gears around the turntable to get a more solid gear train for this high stress axis.  I actually did find a way to fit all the gears into the shoulder for it wasn't braced very well.  I had a limited amount of vertical space in the shoulder for gears.  The next innovation was to put the gears for the rotation axis into the base rather than the shoulder.  This allowed me to make a well braced gear train and put more structure into the shoulder to make it stronger and more rigid.  I already talked about it a bit on @Jim's robot thread.  I kept the motor in the shoulder and moved the gear train to the base.

You can see the medium angular motor in the shoulder and 4 gears driving the turntable in the base.  The power is transferred through an axle through the center of the turntable.  Here is a cutaway of the gears:

 

New version in its entirety:

So how does all this work?  Below you will find a video that combines the new version of the 6 axis robotic arm (https://rebrickable.com/mocs/MOC-109607/glaysche/6-axis-robotic-arm-mk-2/#details) and my 6 axis remote control (https://rebrickable.com/mocs/MOC-94286/glaysche/6-axis-remote-control/#details).  They are connected together using Mindstorms hub to hub communications.

 

You'll notice at the end of the video, the upper hub on the robotic arm crashes hard and stops responding.  It has to be power cycled to use again.  I'm pretty sure this is a bug in TLG's software, perhaps in the hub to hub communication.  This bug has been there for the entire time hub to hub communications have been offered as a feature.  Anyone know a good way to submit a bug report to TLG that they'll actually see and not just go into a black hole?

Edited May 3, 2022 by Glaysche

[Jim]

Thanks for posting this update. I love seeing some progress, after realising you started this topic way back in 2020.

The bug in the hub to hub communication is concerning. We had a brief conversation on Facebook about this. I really hope you will find an answer, because I am likely to run into the same issues. Although I will probably do less hub to hub communication per second/minute.

I can try posting this on the LAN (LEGO Ambassador Network) and see what happens.

[Glaysche]

2 hours ago, Jim said:

The bug in the hub to hub communication is concerning. We had a brief conversation on Facebook about this. I really hope you will find an answer, because I am likely to run into the same issues. Although I will probably do less hub to hub communication per second/minute.

I can try posting this on the LAN (LEGO Ambassador Network) and see what happens.

That would be fantastic.  Maybe you can find the right person to talk to.  I am more than willing to help track this down, running debug builds, doing different experiments or whatever it takes.  Thanks!

[Jim]

8 minutes ago, Glaysche said:

That would be fantastic.  Maybe you can find the right person to talk to.  I am more than willing to help track this down, running debug builds, doing different experiments or whatever it takes.  Thanks!

I will do my best  

[ord]

@Glaysche Wow, I just read through this thread for the first time and inspected the studio model (thank you!) and this is quite the impressive machine you've made.

Using differentials to decouple the joint movements is genius, and I imagine it saves a bit of pain when it comes to programming. What is the backlash like though, considering so many more gears are needed for it?

I read on Rebrickable that you're working on a new version - looking forward to it! Do you plan to omit the 3D printed part and make it pure Lego?