Go Down

Topic: [New Videodemo 31.03.2011] Animating Paper with Nitinol Memory Wire (Read 10166 times) previous topic - next topic


edit: this is not regular nitinol, but nitinol memory wire. as sold here: http://www.imagesco.com/nitinol/memory-wire.html
Sorry about this confusion - I changed the title, I think it should be clear now.


This is another project I am working on. Goal is to create paper which can move about (well, at least a bit)
Here is a video of some first trial runs with nitinol memory wire


EDIT: New video:


An extensive description of the project can be found on my blog:



Currently what we do is really brute force. We just short  a battery to draw enough current to get the cable hot. I believe we completely overshoot the required temperature of 70 degrees celsius. Does anyone here have an idea of how to do this in a more controlled fashion?

(yes, as of now there is no arduino connected, so its not technically an arduino project. however, we will want to controll it somehow, and that is where the arduino eventually will come int.)


Use a PWM controller to switch an FET.

I've done quite a bit of experimenting with Nitinol wire, when the kits first came out in the late '70's.
For simple stuff use a 555 and control the duty cycle with a pot.

You will want to make use of levers!
The wire doesn't contract that much, but it can exert a fair bit of force, magnify the "throw" by at least a factor of 10-100 with a lever and you have a workable actuator.


wow nice! put up some sort of a tutorial and where to get these wires from! very interesting!


Mar 02, 2011, 02:11 am Last Edit: Mar 02, 2011, 02:13 am by retrolefty Reason: 1
I worked on Foxboro Spec 200 strip chart recorders in the 80s. They used a "Nitinol motor" assembly to drive the chart pen through it's range across the slowly moving paper strip. The assembly on the outside looked somewhat like a large servo assembly. In side was about a 6" length of Nitinol wire held under tension with springs. Passing a controllable DC current would force the wire to 'shrink' in length and the linkages and such would convert it to a rotary motion.

Key points as I recall, is that the wire 'shrinks' much faster then it relaxes and I think you can see that in the posted video. So springs were needed to assist the relaxation cycle. Also to be able to maintain a linear relationship of mechanical travel Vs current flow there had to be some minimum DC current flowing through the wire even at it's 0% travel and increasing current would allow smooth 'shrinking' up to it's 100% travel length. They were not very fast acting 'motors' compared to motor driven servos but the thought was that they would have better long term reliability Vs motor brushes, etc. For hobby purposes keep in mind that these can be real current hogs (we are taking amps in the larger gauges) for all but the smallest gauge Nitinol wire. They can supply pretty strong forces in the larger sizes, and can be made to have linear travel Vc current, but they are no match for the efficiency of gear train driven motor servos.

So possibly a cool thing for artistic applications, but not a real practical controlled motive force driver.


Mar 02, 2011, 05:13 pm Last Edit: Mar 02, 2011, 05:29 pm by fkeel Reason: 1
first of all, I posted this under a somewhat misleading title - sorry about that.

The material I am working with is *not* regular Nitinol, but something called "Nitinol Memory Wire". Nitinol shrinks (or expands? i get it confused) when heated and then expands (or the other way round) when it cools again.

Nitinol Memory Wire has two phases. At room temperature you can freely shape it to anything you want and when you heat it, it will go back to its original shape. When it cools, it stays in that shape. So for bidirectional movement you need a counteracting force. Memory Wire is the material used in the above video.

I bought mine from a NYC based company called images. They sell all kinds of really cool nitinol-like products (they also sell actuators very similar to what retrolefty describes). Find them at www.imagesco.com

thanks, retrolefty & cyberteque for your input.

I have also come to the conclusion that nitinol is not really practical. I was hoping to get a setup which could animate flexible displays. My first thought was that the main problem would be the temperature changes which need to be taken into account. However, as retrolefty says, there is also the additional obsticle of really high current requirements.

btw @ retrolefty. You mention, that it is slow. However, if you get your setup right, it can be extremely fast. Ever tossed some into boiling water? I did it and the shape change was so abrupt it ricocheted back and flew through the entire room...

The observation of one direction beeing strongr than the other is possibly due to the fact that I just heat it up using a bunzen burner. I believe what happens is that I overheat it and slightly damage the material. Another effect which we have to deal with is that sometimes its physically impossible for it to return to its original shape - however I think the temperature of the cable (as we have no regulation on that what so ever) may rise to temperatures high enough to set it into the new shape, which in turn again will reduce its strength, making the cable weaker and weaker with every trial.

does anyone know of a sensor which could measure the heat? which is fine with temperatures abruptly changing from room temperature, to potentially several hundred degrees celsius?

@ cybertque ... I was playing around with PWM for led's and was thinking about trying something along the line of what you suggested. I will see if I can get around to trying that one of these days.

I played around a bit with regular nitinol (the dynalloy stuff...). My main problem with it is the ridiculously small gouge on it, which makes it really hard to handle ... its like trying to build something out of hair... when you where working with nitinol, do you remember what gouges your nitinol had? how did you manage the filigranness of it all?



Once my project is done and if its cool I'll do an instructables on it or something. As of now, all the info I have can be found on my blog (fkeel.blogspot.com) and www.imegasco.com also has some good documentation on it. I'm happy you are interested in it though, and if you have specific questions feel free to ask or send me a pm or something.


For the fine gauge wire I used 4BA bolts & nuts on model aircraft control horns, this works well to insulate the circuit.
I wrapped the wire around the bolt a couple of times, did a dodgy knot like the one you use on a fishing hook, then tightened the bolt & nut, the electrical connection was a piece of brass strip with the lead soldered to it.

The larger gauge "biowire" was pretty thick, about the same as nichrome wire.


hm... I did not have time to incorporate all the tips I got here, and in the electronics subforum. However, Thank you for your input with that.

I am for now dropping the project, but I made a short video to show what we did.

So take a look:


still no tutorial - as I still am not really able to controll them to the extent I would like to.

anyway, I think what we did is still pretty cool, would be interested what you guys think.



Go Up