threedigits writes:
"The technology to build giant robots is one bit step closer: a group of international researchers have published a paper describing a method to create artificial muscle fibres. The cool thing is that they are about 100 times stronger than biological muscle tissue, and you can try it at home! The basis is polyethylene or nylon string, AKA fishing line. A great video is available on Hack A Day."
This discussion has been archived.
No new comments can be posted.
Researchers Create DIY Artificial Muscles
|
Log In/Create an Account
| Top
| 11 comments
| Search Discussion
The Fine Print: The following comments are owned by whoever posted them. We are not responsible for them in any way.
(Score: 4, Funny) by VLM on Sunday March 02 2014, @09:30PM
"artificial muscle fibres"
White meat or dark meat? Taste like chicken?
(Score: 1) by rudolph on Monday March 03 2014, @07:45AM
At first it tastes like nothing, but after chewing on it a while it begins to taste like blood.
(Score: 3, Interesting) by frojack on Sunday March 02 2014, @10:13PM
There is a non-trivial problem when trying to use a torsion inducing twisted element in a structure where you want to harness only the tensile capabilities.
Imagine your bicep being a twisted element. You would have to anchor both ends to the humerus, unlike the situation we have in the human arm [learnbones.com]. (Other wise you would your forearm tending to windmill from the elbo.).
Of course, once you anchor both ends to the humerus, all you can do is stress the humerus. You aren't going to be doing any lifting.
Any structure you imagine to employ this type of artificial muscle must prevent the torsion from inducing any twist in tensile element.
In the video, he used a long lever anchored to a support structure to prevent the fish line from untwisting.
In a real machine, you's probably have use some sort of channelized structure to prevent untwisting. (Think of a hydraulic cylinder [dreamstime.com] with a square cross section instead of a round one).
That would be a high friction device, you'd have a lot of power lost just overcoming the built in friction.
I'm not convinced this method is any better than current forms of liner drives, which are almost always rack and pinion, screw thread, cable and drum, hydraulic, or pneumatic cylinder devices. We have quite a few basic machine elements that fill this role.
No, you are mistaken. I've always had this sig.
(Score: 2, Insightful) by Boxzy on Sunday March 02 2014, @10:49PM
I'm pretty sure simply using mirror pairs of twists would remove the effects of torsion.
My problem with this is that heating and cooling will be quite slow, you wouldn't want to try designing a robot that needs reflexes with this, I imagine each contraction and relaxation would take on the order of seconds.
Go green, Go Soylent.
(Score: 3, Interesting) by frojack on Sunday March 02 2014, @11:17PM
Good point regarding the pairs of opposite twists.
But if you had a flexible heat wire embedded in the twist, (or a core around which the twist is wound), you could probably apply that heat fairly quickly, relative to size of the over all device. Embedding a small amount of metal in the fish-line might make controlling it with microwave heating fairly quick.
But yeah, the heat gun clearly isn't going to work.
No, you are mistaken. I've always had this sig.
(Score: 1) by Boxzy on Monday March 03 2014, @09:12PM
No matter how much energy you can put into heating super fast, cooling is always going to be slow. As many of us are computer engineers removing heat is something we all are familiar with.
Top three cooling methods in order of difficulty:
1. Air cooling, far too slow for this application but probably low power and weight enough for mobility.
2. Peltier cooling, Too high power requirements, quicker but how are you going to ensure a good contact with a coil that increases in size when hot? Highly impractical.
3. Good old water, will retard heating just as it is good for cooling. Still slow.
Are there any other options?
Go green, Go Soylent.
(Score: 0) by Anonymous Coward on Sunday March 02 2014, @10:57PM
Or you could mount it in such a way that it's free to twist, seems to work fine that way, at least from what I could see from this video http://www.youtube.com/watch?v=-Fh65grtjS4 [youtube.com] from hackaday.
Something like a fishing swivel should work just fine for that application.
I don't know how much that would reduce the power of this thing, but might be worth a try?
Anyway I found the aspect of it being heat driven strangely appealing and the fact that the material is readily available and cheap should open up for more research being done into wether it could prove benefifical over more conventional actuators for practical or economical reasons.
(Score: 2) by frojack on Sunday March 02 2014, @11:20PM
Very cool.
That's an interesting twist (see what I did there?) on the original video, as he annealed the material first.
No, you are mistaken. I've always had this sig.
(Score: 1) by NovelUserName on Sunday March 02 2014, @11:08PM
In the HAD video where the energy source is heat application, there doesn't seem to be much torque applied. From the paper abstract(I don't have access to the full document at the moment) it sounds like you might be able to drive these by applying torsion and the end. In that case you could presumably put the fiber in a sheath and twist against the sheath rather than the insertion point of the muscle.
Either way I think a bigger problem is sourcing power: The claim is that these things have a mechanical power output per mass on par with a jet engine, so you either need a heat source large enough to support that kind of energy transfer, or a torsion motor with the same power output. When you consider the combination of twisted fiber AND the energy source, the size is probably much less impressive than noted by the research team (The HAD video shows a heat gun that probably weighs 10000x the fiber that they are demonstrating).
That said, this is a cool way of driving linear motion- I may have to play with it some time.
(Score: 1) by jmoschner on Monday March 03 2014, @02:01AM
While early tests are using things like heat guns, heating elements could eventually be woven betwen the fibers.
(Score: 1) by MickLinux on Monday March 03 2014, @11:16AM
Why wouldn't you use a pair of muscles, one left-twisted, one right twisted? Or actually, a quad, two and two?