Replacement for Pneumatics and Linear Actuators

[mister923]

Hi Everybody,

Over the summer I have been interning at a startup (because of my time on an FRC team) that is making a replacement for pneumatics and the classical linear actuator that team might be interested in using!

Watch a video of them, Cyclone Muscles, in use in a 2014 bot below!

https://www.youtube.com/watch?v=mgjhWmUQaqY

The company is Rise Robotics in Boston.
http://riserobotics.com

[Andrew Schreiber]

Originally Posted by mister923

Hi Everybody,

Over the summer I have been interning at a startup (because of my time on an FRC team) that is making a replacement for pneumatics and the classical linear actuator that team might be interested in using!

Watch a video of them, Cyclone Muscles, in use in a 2014 bot below!

https://www.youtube.com/watch?v=mgjhWmUQaqY

The company is Rise Robotics in Boston.
http://riserobotics.com

Color me interested, got a link to learn more?

[mister923]

Check out the website: http://riserobotics.com

We have a beta test running if you are interested!

[AdamHeard]

Does the device only apply tension force?

[magnets]

This is a very cool concept. I can see how it would be useful in moving manipulators and catapults in a way much simpler than some teams currently do.

A few questions/comments:
Can in both push and pull? I'm having a hard time visualizing.

What sort of rods are used, and what type of linear bearings are used?

What exactly is the spring like thing at the end?

How much force/twisting/misalignment at the ends can be tolerated?

I'm a bit confused by the specs. It says 179 watts is the max power, which is exactly 110% of the power required to lift 778 newtons at .21 m/sec. This makes sense because the listed efficiency was above 90%.
However, the rated load current draw at 12 volts is 21 amps(252 watts), which is not equal to 179 watts.

I'm trying to get an idea on how the compares to a ball screw/lead screw set up. I'm pretty convinced that for medium loads, this could be a heck of a lot better.

[mister923]

The Cyclone Muscle is biologically inspired and like your muscles it is a PULL mechanism.

[mister923]

As to the rods, and springs I am a bit confused as to what you have in mind? Could you clarify?

I will have a better answer and updated specs for the current model soon.

[Jon Stratis]

This is a very interesting mechanism... my biggest concern is with how exposed all of the workings are - it seems like it would be fairly simple for something to get in the way of the motor track, or for something to impact the string/cable, possibly breaking it or knocking it off its guide. Are there any thoughts to having a protective shell that goes around the whole thing?

[mister923]

The next version of the Cyclone Muscle has been designed with a case and that is the version that will be sold. The version you see in the video was an early prototype.

[AdamHeard]

Originally Posted by mister923

The Cyclone Muscle is biologically inspired and like your muscles it is a PULL mechanism.

I'm a bit confused about it's advantages then. If a lead screw mechanism were designed to be pull only, then it could be very compact and likely substantially cheaper than the $191 seen here. This takes up a relatively wide cross section, and to cover the vulnerable innards would increase that.

You'd have a tough time selling me on the reliability of the cabling (not unspooling ever, etc...) as well.

What type of printing is done on it? I'd have concerns about the strength of the eyelets too depending on motor ratio.

[evanperryg]

Originally Posted by AdamHeard

I'm a bit confused about it's advantages then. If a lead screw mechanism were designed to be pull only, then it could be very compact and likely substantially cheaper than the $191 seen here. This takes up a relatively wide cross section, and to cover the vulnerable innards would increase that.

You'd have a tough time selling me on the reliability of the cabling (not unspooling ever, etc...) as well.

What type of printing is done on it? I'd have concerns about the strength of the eyelets too depending on motor ratio.

If these could move faster than a lead screw, I'd certainly consider using them. Really clever design, with some more improvements they could become a valid replacement for pistons in some applications. (though they really should increase the preload on that shooter )

[AdamHeard]

Originally Posted by evanperryg

If these could move faster than a lead screw, I'd certainly consider using them. Really clever design, with some more improvements they could become a valid replacement for pistons in some applications. (though they really should increase the preload on that shooter )

That's a gear ratio and/or pitch decision. You could have a lead screw go pretty much any speed you want.

[magnets]

Originally Posted by AdamHeard

I'm a bit confused about it's advantages then. If a lead screw mechanism were designed to be pull only, then it could be very compact and likely substantially cheaper than the $191 seen here. This takes up a relatively wide cross section, and to cover the vulnerable innards would increase that.

You'd have a tough time selling me on the reliability of the cabling (not unspooling ever, etc...) as well.

What type of printing is done on it? I'd have concerns about the strength of the eyelets too depending on motor ratio.

It is an efficient mechanism (at least when there are no side loads, those linear bearings don't seem to be much) and it can handle 175 lbs, though it'd probably need a decent amount of support.

I do agree with your concerns though. I wonder how long the plastic gears and string will hold up for with the sort of shock loads we see. I'd also be a little scared of using this to drive something like an over the bumper collector.

I don't agree with the site about lead/ball screws(or most of the comparison page for that matter): "However they suffer from sliding frictions causing these devices to be inherently slow and limited in operation." Ball screws are often over 90% efficient and can go faster than the cyclone and have no sliding friction. They do require fancy thrust bearing setups though.

While it's definitely a cool product, I'm not sure what it would provide over a normal winch with slides. Having the carriage move back and forth is neat and lets you get away with less gear reduction, but I don't know if it would play well in FRC.

You can go much faster with a multi start lead screw:
https://www.youtube.com/watch?v=QQ-VOW2GIyY

[mister923]

It should be noted that the model that is on the website and video are prototypes. The comparison of ball screws to the Cyclone Muscle on the website is not the most comprehensive as it does not cover every aspect and different setups like mentioned.

The Cyclone is a entire linear actuator system, including sensor output, and is significantly lighter than a ball screw drive train that could be deployed in a robot for FRC.

As for the reliability of the cabling it certainly surpasses the requirements of the demands that would be put on the system during a season.

The Cyclone will tolerate 3 degrees of angular misalignment.

Below are the revised specs
The Cyclone's motor currently spec'd to run at 1/3 of stall torque which corresponds to 144 mNm, which at 12 V outputs 165 W of power with about 26 A of current.

As to efficiency, 80% is a better approximation for the efficiency of this version as it includes a gearbox. The Force output for the shipping model is 140 lb = 615 N, P = F*V/e = 165 W.

Then revised specs, 165*0.8/615 = 215 mm/sec


I hope that answered some of the questions.

[asid61]

Pretty cool. I don't think it can beat a balll or lead screw in terms of weight and efficiency, however, ball screws are extremely expensive. This saves some weight and looks like it is very cheap and easy to build.
I am worried about the strength of the cabling. What is the tensile strength of the cable?