Introducing Andrea Aglietti

Viewing 10 posts - 1 through 10 (of 10 total)
  • Author
  • #487

    Hi I’m Andrea from Italy; with my girlfriend we have started building a “Thor” copy following the original project as her final thesis in electronics and communication at University of Florence. Our goal is to end up with a functional prototype on which she could work on the electronics and on the control software. The arm will be used as a part of a NATO project by her professor. We have printed all the parts using a professional photo polimerization printer at university and a tevo tarantula 3d FDM 3d printer that we have at home.We’ve started working on the electronics making our own case for the original electronics, we have then made cables to reach the robot and now we are working on the assembly process of all the printed and no printed parts. We have succesfully build many parts and we have tested the art56 that is working fine (only some little issue that we have enconutered during the building) but now we are struggling with the art4 gear, it seems that the DIY plastic ball bearing doens’t fit tight enough so the axis of the art 4 is pushed away from the right position when the stepper motor start to run ending in no motion and stall of the mechanism.Hope to get some information and feedback from other users that already have built succesfully the robotic arm.
    Thank you so much
    Andrea Aglietti

    Angel LM

      Hi Andrea!

      First of all, welcome to the Thor Community! 🙂

      The DIY bearing of the 4th articulation is bit tricky to print because of 3D printing tolerances. The current piece fits with my printer tolerances, but yours could be different, or maybe we are using different plastic balls. The solution is to modify the bearing pieces, increasing the torus radius. As it could took several attempts I would recomend you to design a simple bearing with two cylinders and tune the radius of the torus, and when you got the good one, modify the original radius and print the art4 bearing pieces 😉

      If you need some help with the process do not hesitate to ask!


      Hi Andrea, Welcome to this site. It a great community for help in building this open source robot arm.
      I have sent you the ART4 TramsmissionColum design in a step file that I created from Fusion 360 for you to also try. It will help you print all parts on your printer.
      RegardsMartin Barr.
      P.S I am also looking and designing an Idle Gear for the other side of the transmission column, this is based on the redesign that was done by Danny Van Den Heuvel. I am printing now and going to assemble to test later this week.


      Hi Angel,
      ok i will try to increase the torus radius, now I have too much play I’ve tryed to print your original design with both a professional 3d printer and my tevo tarantula but the result is near the same I have too much play and another problem I have is that the motor isn’t fixed strong enough and it is able to move a little when it try to rotate resulting in skipping of the shaft.
      I’m also trying the ball bearing design of Martin. I’ll make some test when I’ll have some free time and I’ll give you some feedback.A question: do you have a skype account?, It would be great to have a talk so I can also show you my issue in a videocall.
      Thank you very much.
      Andrea Aglietti


      Hi, ;
      I’ve finally got it working, I’ve put some bigger balls into the bearing and know all the play that had before is disappeared and the mechanism is working, I only notice I have some backslash due to the gears but I will work on that later, for now it’s ok.
      Thank you,
      Andrea Aglietti

      Angel LM

        I’m pleased to hear that! 🙂


        Hi Angel,
        the robot is finaly working, all the things mounted and cabled and so on, electronics is ok and everything is fine. Now I’m making a simple interface in processing to control every joint of the robot separately; this is my first attempt but after I would like to implement a 3D model on the PC with the inverse kinematic in order to have a direct control of the robot in real time.
        So I have two question:
        – 1 with the grbl firmware you’ve uploaded on github and that I’m using is it possible to operate in jog mode? it would be great in order to have a joystick control style.- 2 do you have any piece of software already written and with open source license that you could provide to control the robot?
        I’ve made my very first try using a Gcode sender (universal G code sender) and now I’m moving to the solution made in processing but maybe there is something already made that works fine and is open source.
        Thank you very much,
        Andrea A.

        Angel LM

          Hi Andrea!

          Yay! Another builded Thor! 🙂

          Regarding the first question, I don’t really know if the GRBL version that I adapted years ago can work in Jog mode… GRBL v1.1 have that option. I’ll check it and see if I have to migrate to a newer version of GRBL!

          About the second question, I’m currently making a basic application to control Thor. It’s called Asgard and it’s available on this Github repo. It’s written in Python and Qt and it’s OpenSource, of course.
          At this very moment the aplication does nothing. (I only designed the Graphical Interface) but I hope that I’ll have a basic version of the program in the following days/weeks (just forward kinematics). Maybe I’ll open a thread here to keep the community updated about it.

          Best regards!
          Angel LM


          Hey Martin, ;We are currently having the same problem with this art 4.Could you also send me the modified stl file to my email please?Thanks in advance.
          Best regards,Thalles Machado

        Viewing 10 posts - 1 through 10 (of 10 total)
        • You must be logged in to reply to this topic.