Ultra-compact 6dof Linear Platform

Howdy folks!

I've been cruising around the X-Sim discord and finally figured it would make sense to post some progress on my project here in a thread. I want to start right off the bat by crediting the builds that I have spent hours reading and drawing inspiration from, as well as the creators who are sharing their knowledge with me and the community. Those people in particular are Wickie and SilentChill. Without their builds and willingness to share I wouldn't be able to get as far as I have so quickly. Thank you both!

The project is for a self-led engineering co-op because I couldn't get a job... all the postings were for outside of my city and I have a 6 month old so that was a no-go! It's a happy thing though because now I can put full-time effort into my project.

As a full disclaimer, this project IS funded by an external investor who expects there to be some monetization strategy at some point, so there may be some details which I can't share. HOWEVER; this stage of the project is entirely open-source and doesn't really feature any innovations that I would want to restrict the X-sim community access to, so hopefully my work here helps people as much as those projects who have come before me have.

The idea for my project is pretty straightforward: motion sims are big! dangerous! expensive! and TOO MUCH FUN! Not everybody has space for a permanent installation of a 6DOF system and might have to pack it up if they need to reclaim whatever space they could have a rig set up in. So the plan is to make something that can bring 6DOF to more people.

I already have a few innovations in mind which I can't share too much of here, but suffice to say it requires some custom engineering and fancy parts. Those two things = money so I'm starting the project with a Phase 1 where I build a platform based off the INCREDIBLE work of those here on the X-sim forums as a way to find the parameters that provide the performance I perceive will put enough pep in peoples pedals without peppering their properties with parts. After I've gathered that data it will be MUCH easier to justify spending more cash on the custom designs.

Okay, walls of text are boring AF so I have to oblige with some photos. ODrives, 140kV BLDC motors, AMT102-V encoders, six 300mm 1610 Ball Screws with FK12 + deep groove ball bearings, and a 3D printer for all my housings and parts are what I have so far! Oh yeah, and a $20 HP MS200 running LINUX for easier work on the ODrives. I hope that's enough to placate the audience for now

I should add; I've picked up a parts washer to clean the shipping grease off the ball screws and will be relubricating them with a thinner oil. It's not good to run them dry, but the loads these are going to see are a solid 15x lower than they are rated for so they don't require the same level of protection. I have a spray lube which is safe for food contact so it should be acceptable for minor contact with carbon fibre and nylon.

good luck with the project and i hope it leads to something bigger and better in the job department

what type of motors do you use / voltage etc

Thanks Gadget! This will definitely be the jewel of my portfolio for now.

I'm using 6354 motors at 24V. That will get me ~3000 rpm, which with a 1610 will result in 500mm/s movement. The motors are a bit skimpy on specs but they will operate around 600W max power outputting ~1.7 Nm of torque, which should get me around 2G of force. I have three Meanwell RSP-750-24 PSUs that just arrived from UPS so I'll bring those about to my workshop. I was going to get 1000W PSUs but decided to save a few dollars... unfortunately the 750W versions don't have current sharing like the 1000 however from my research these will be more than adequate to drive the ODrives. Everything is overspec'd for now and will be focussed into a much tighter spec with the next iteration.

These are all theoretical numbers for now, I'm sure the load performance will be less. I've got

June and July were a wild ride. Now that we're halfway through August I thought it might be a good time to post a little update on the project.

Since the first phase is to build a 6DOF to validate the actuator performance I haven't designed anything proprietary. In fact, the actuator design I have gone with is quite similar to one SilentChill (thank you again for sharing so much detail of your build). If you see something you like I'm happy to share info about it!

All parts are printed from carbon reinforced Nylon and annealed at 80C to improve hardness. The four shafts covered in the protective sleeves are hard anodized aluminum linear shafts from IGUS; 3/4 will be replaced with non-precision shafts of equal diameter, probably made of carbon fibre, in the full prototype assembly.

There are a few components missing, chiefly the ball nut --> drive cylinder sleeve, limit switches, and end caps. However, the majority of the actuator is completed! I calibrated my printer for IGUS i150 filament to make a bushing out of, however some more testing needs to be done before the parts fit together nicely. And since this is just for validation of the speeds, acceleration, etc. I am probably not going to worry about it. That said, it prints quite nicely (shown below is a flow calibration cube of the IGUS filament).

Next steps will be to get the ODrive configured for one actuator and perform some tests.

Recommend that you use the required grease and not thin oil. They will run quieter and wear better. Once any bonding is done, never had a issue with it worrying any carbon fibre or nylon. FI cheap screws also wear in and become quieter after a while.