For those of you who don’t know, i am building a large scale 3d printed robot arm. The main structure is made of carbon fiber rods, with many bearings used in the joints in order to constrain the motion of the robot. My design takes inspiration from a class of robots known as parallelogram robots. These robots make use of large linkages and joints in order to keep all of the motor mass tucked close to the base of the robot. A major aspect of these robots is the static orientation of the head. The head will always be orientated parallel to the base due to the dual-parallelogram structure. This makes these robots ideal for precise, palletizing as well as pick and place tasks.
However, there is one problem with my design. The gearboxes suck. I have been using worm-drive reducers for their simplicity, However, they are inefficient, non-backdrivable, and fail under heavy load. I am going to be switching to a belt-driven reducer. The simple 4-stage mechanism that you saw in the thumbnail allows for a 51: 1 reduction ratio. If tensioned properly and rigidly assembled, the belt reducer could have almost zero backlash. Given that it costs nothing compared to similar performing alternatives, this design is very promising. I will be constructing the initial gearbox prototypes soon, so look out for updates. I cannot continue on the software development until I have new actuators, as my current ones have just proven themselves to be incapable.
Soon I will make a post showcasing the actuators in more detail, but I do not currently have access to the 3d model at the time of me writing this. It is exciting to be so close to a working arm.