Idea for a cable-suspended 6DOF Stewart Platform

Hi everyone.

I made a post earlier with an idea to use springs to support a seat so cheap wiper motors could be used for a 6 DOF Stewart Platform and still have high speed and long travel. This idea kind of went to bust, as standard compression springs couldn't be used with the lengths I needed (the math just wouldn't work out). I then spent an extremely large amount of time trying to figure out a way to use a pneumatic system, controlling speed and positioning of a pneumatic cylinder by regulating the air pressure going into the cylinders. This also turned out to be a bust.

Long story short, I figured out a way to use cheap wiper motors and still maintain high speed and long travels. Basically, I just use a simple chain driven system. Here is a picture of a simple sketch I made:

The wiper motors (I am looking at the motor here) have a roller chain puller attached to them. As per my sketch, using a 1.75" dia. pulley results in 10"/s speed, and 100lb of thrust. The roller chain connects to a steel cable, which loops around a pulley attached to one of six vertical posts. The cable then attaches to one of 6 connections on the seat. The other end of the chain has two options: it either goes through a chain guide, which is a square opening with rollers on each side, and then connects to a steel cable, or I could just mount the motor on a thrust bearing so the motor can rotate, eliminating the need for the chain guide, allowing the chain to just connect to the cable then to the seat. My goal here is to eliminate the chain from binding or slipping off of the pulley when the cable is pulled at an angle.

I plan on purchasing my steel cable from here and my pulleys/chain from here. I will be using the Sabertooth 2X12 motor driver here, and I will be using an Arduino (or arduino-based controller that I will make from a higher powered microcontroller).

I am looking for any feedback regarding this design. I have built a small prototype using wood dowels and servo motors, but I need to figure out a different power supply as I can only power 5 of the servos. I plan on configuring this design to work with simraceway, and watch it move on my desk as I play the game just to verify the motion works as I hope. I look forward to any addition insights.

Nice idea be an interesting build. Keep us all informed, has potential.

You can run more than 5 servos on an Ard, remove the power , ie power the servos from a psu - 5-6v instead of the Ard or us a proper Rc controller , Pololu make some good units, Photos of model would be a nice addition.
Yes Simraceway will work for you here in SimTools.

Thanks! I actually am using a separate power supply (just a phone charger that I stripped) with roughly 0.4Farads worth of capacitors and I'm still having issues. I have a PSU laying around, I just need to get to wiring it up again.

Are there any examples of the axis settings for a 6DOF setup laying around? I haven't been able to find any, and it'd be nice for a "jumping off point".

I'm also still looking for a seat, any recommendations?

in your case (suspended design), I would say "light weight"
and personnaly, "comfortable" as well!

fiber rally seat could match these 2 conditions...

Need to feed the servos with at least one amp, I use a rc BEC in the power loop so I can use a 12v 2amp psu to feed the servos with a contant 5v's.

On seat I agree with @RacingMat , lighter the better, I use plastic baja seats , just under 3kg with very little padding, not that comfortable but I build em to drive , lol you don't see a big arse comfy seat in a real race car. So I keep it real.
Am hour in the sim feel like an hour in a real race car, love it.

Hi Taylor interesting design, you may need a chain tensioner to ensure tension is the same on each wire as they move. That design may also allow some unwanted twist in the frame, which may or may not be an issue. FYI from my experiments, the output from Simtools can not exactly relate to a Stewart Platform (but it is close). Unsure how those errors will effect your rig. Building a working model as a proof of concept is also a good idea, one based on your proposed design.

Hi Taylor, how did you get on with this. I am trying to do the same thing