Question How about an inverted hexapod with cable winches?

I wonder why nobody had the idea to build an inverted hexapod motion rig that has a pod hanging on six steel cables and winches bolted to the ceiling. At least I haven't seen one, yet. I think it should be much easier to build as you wouldn't need expensive parts like ball screws and linear bearings. All you need is geared motors and drums to wrap the cables around. Even the U-joints are unnecessary as the cable easily bends in all directions.

Of course, there are some limitations. As the cables can only pull but not push you can't generate negative g's and you can't roll or pitch faster than the pod can tilt when fixed on one side and free-falling on the other. But I think those are quite extreme conditions that are rarely ever needed.

On the other hand there is nearly no limit to the stroke of winch actuators. If you have a high hall, barn or hangar you can easily make the heave 5m high or more which could be used to produce some serious g-forces or even near zero-g for fractions of a second.

Can you provide a link to an example? I've found some discussions about winch motors but AFAIKS they use only the gear motors but not the winch itself.

Edit: found a small demo project.

I think some of the issue might be precision of motion. When you're relying on gravity to do half the work and you have flexible cables, you aren't really constraining the rig the way you think you might be. I think you're likely to get the whole "pod" swaying over time as you pull on the different sides. I like the idea in general though and I think you could probably make it work better with some extra cables to control things. Maybe almost like spring dampers that help tame out any resonance in the system.

There are surely some limitations. For example, the circle of the pivot points on the ceiling has to be greater than the circle of attachment points at the pod. This way the cables always pull a bit outward. Otherwise it would be like a playground swing. You also have to limit the (downward) acceleration so that the cable tension stays positive (no slack), maybe 0.3 to 0.5g min.

But I think if you use steel cables and some sort of "thread" on the drums to force a single layer and clean wrap then positioning accuracy should not be a problem.

If the vertical stroke is greater than ~1m it would be a good idea to add a safety belt from the pod to the center point at the ceiling. A centrifugal clutch can limit the falling speed in the case of a power loss or gear failure. For <1m distance to the ground dampers made of styro foam should be sufficient to avoid injury.

It should work well for a fight sim

But you are limited to -1g when working with gravity

Probably not suitable for a driving sim

Might be a large expensive build

You really think you need negative g-forces for a driving simulator? I thought a car could free fall after driving ofer a large bump or steep hill. In this case you're effectively weightless for fractions of a second which could cause indeed problems to the winch hexapod. But you never need negative g.

Well, you could if you really have large spoilers at high speed. You can also fly an inside-out looping or an inverted barrel-roll. But that cannot be simulated properly with a standard (non-inverted) hexapod either.

I thought about this as well along with a lot of other people have it too I think the problem with this design is a backlash hard to keep the cables rigid enough to do the push. But actually I think it could be done with not cable but like a sort of a an interlocking linkage type system that can bend and then lock in place it will

I did some calculations. The problem is indeed the "playground swing" effect. The rig is controllable quite well as long as the angles of the cables stay close to a regular triangle (60°). But the longer they are the smaller the angle gets. If the pod is far away from the pivot points at the ceiling the cales are nearly parallel. In this case a very small amount of slack is enough to loose control of sway and surge. You'd have to limit the acceleration to much lower values than 1g.

So the best solution to keep dynamic control in all situations would be to add 3 cables that pull down to form 3 tetrahedrons instead of 3 triangles. So this turns the hexapod into a nine-a-pod (enneapod?). Well probably too much effort...

Seems like the bandwidth and stiffness will be low (things I want to be as high as possible for racing). Maybe ok for flight sims, I suppose it depends on your priorities.

Like I said I think a cable type system made out of linkages break apart linkages that will stiffen up on a when they're straight but will also allow them to break away as they go around the pulley I think something like that would work I'm sure they have some out there but you may have to design one but it seems to me like a linkage type cable would work

Yes, for very long cables bandwith might be an issue. The cables are basically like guitar strings. A 10m cable can have resonances in the ~10Hz range. This might feel "bouncy" when trying to simulate a rather stiff gear of a racing car.

Ah, the pod in the video also has cables pulling downward. Are they spring loaded or actively driven?

What do you mean with "linkage type cables". Something like a chain? I think the stiffness (force divided by stretch) of a steel cable is not much different from a steel chain. But they have more vibration caused by the chain links hitting the pulley teeth.