Experimentation will have to verify this but I think this setup will be wobble tolerant. The large wheels should be able to shake back and forth normal to their surface and the separation between steppers and the tool should remain relatively constant. A typical delta robot would articulate at the ground joint but I am articulating from the mid-joint. (To be technical, a typical delta cantilevby nicholas.seward - Delta Machines
Very true. However, I am dreaming of insane accelerations that will generate forces that overshadow gravity.by nicholas.seward - Delta Machines
QU-BD has an 8" for $20 but it is also 24V. You can use two PC power supplies or get a regular power supply. I am sure that QU-BD will make custom ones for you if you are going into production.by nicholas.seward - Delta Machines
Interesting idea about the screw. Depending on arrangement, it could impede on the work envelope but I think that can be addressed. (I only worry about backlash.) I think it is a very workable idea. I am going to experiment with the belt/string drive to see if it is feasible. Speed is more important to me than small steppers. In fact, the Nema17 motors are 15% cheaper than the Nema14 ones Iby nicholas.seward - Delta Machines
@ShadowRam: If I print them solid (which I won't) the top pulleys will weight 40g each. However, I have not worked to optimize their size yet. I was trying to get a proof of concept down on paper so I could better think about the next steps. Now that I am focusing on the torque needed, the design as is can produce up to 4N forces from each arm. A quick calculation shows that my Printrbot is pby nicholas.seward - Delta Machines
Your right. There are 22 bearings total. However, they are all cheap inline skate bearings except for the 2 on the tool. (Note: my Printrbot has 6 rods, 10 linear bearings, 2 threaded rods, and 5 bearings.) It would be very hard to make without a 3D printer. Luckily, I have a 3D printer. (You would have to make/get a RepStrap to be able to make this RepRap. Standard RepRap plans are usuby nicholas.seward - Delta Machines
Oh, I see. I like that. I am going to try normal bearings to see if that works. If not, I will try this out. Thanks so much for taking the time to sketch it up.by nicholas.seward - Delta Machines
I am not following. (I am not sure if you are saying this but I could instead of using one bearing for each connect point I could use two and increase my rigidity.) Side note: there is only really a need for two bearings. The arm that attaches to the middle can attach directly to the tool holder. For the animation, I saved time by making them all the same but when I get ready to print I willby nicholas.seward - Delta Machines
@VDX: Possibly. However, those are ball bearings on the tool. I think the low level of friction will make this work. I will try a mock up and see if it fails miserably. In my mind (an admittedly scary place), rigidity would increase at first as the distance between the tool mount points increase. Do you have any ideas on how you would sure up the design?by nicholas.seward - Delta Machines
I am pleased to present my first design iteration. Features: Low vitamin count Low friction (Nema14!) Low center of gravity Low weight Can fold flat with some quick disassembly Only 2 bearings will need to be special ordered Printable on a 150mmx150mm plate Around 8" diameter by 8" tall work envelope 9 unique printed parts Watch the video and let me know what you think. (The belts are actuallby nicholas.seward - Delta Machines
Here is the picture of my design that has up to a one inch diameter dead zone in the middle of the platter. I love the simplicity. I don't love the hole. (This is a functionality drawing. It is not meant to reflect how I would actually build it.) I am scrapping this idea. I am still in my brainstorming phase.by nicholas.seward - Polar Machines, SCARA, Robot Arms
Assuming you have perfect alignment, you will be able to reach the center. However, you will have limited speed options as you near the center. On a design I am considering, the arm swings up, down, left and right along with a rotating platform. This means that there is a small region in the middle of the platter that is unreachable. (I think I am going to scrap the idea but I am still curiousby nicholas.seward - Polar Machines, SCARA, Robot Arms
@unlimitedbacon: I have been looking to design something like this with one arm. For acceptance reasons, I would like it to have the same working envelope as a Mendel. The problem is that the practical work envelope is donut shaped. Are you planning on having a no go area in the middle of you plate?by nicholas.seward - Polar Machines, SCARA, Robot Arms
I understand why most don't print the frame. It would take a very long time. I think for most the cost of components is less than the printer time is worth to them. That said, I think we need some well engineered printable frames to give people the option to reduce the vitamin count. What I don't understand is why is pvc pipes underutilized. It seems like we could replace aluminum extrusionsby nicholas.seward - Developers
I should have caught that. Harmonic is a much better pick for this application. The world also needs more Harmonic Drives! I have never seen a actuator like that. I am sure a plastic version can be made. For a simple solution, the z-ish axis could a platform with four(three would be enough) longish parallel bars connected to all the corners. These bars could be perpendicular to the z forby nicholas.seward - Developers
@MattMoses Great Stuff. I have read through a lot of that stuff. Somehow the printed x/y rail escaped me. Very cool. My delta design was something that I had to get out of my head. You know those ideas that you get that you know are probably bad but you have to address them anyways. My thought was that mounting of arms and motors could be easier if that all occupied the same plane. Additioby nicholas.seward - Developers
Cheapskate requires quite a few vitamins. I am dreaming of printing everything but the motors, build plate, and the extruder. Why not have the steppers snap in instead of using screws? Etc. I really have no problem with linear motion (Linear motion is great. There is a reason it is so prevalent.) other than that it requires a lot of vitamins. Additionally, innovations can come from evolutiby nicholas.seward - Developers
I have been rolling around tons of ideas in my head on how to make all the steppers static and get rid of linear rails. I am positive that there is a simple solution out there that can drastically reduce the vitamin count. Rostock, RepRap Morgan, and Ultra-Bot are pretty inspiring. I love stuff about each of them but they are still using rails. I came up with a horrible variation of a delta rby nicholas.seward - Developers