elementcollector1
International Hazard
Posts: 2684
Registered: 28-12-2011
Location: The Known Universe
Member Is Offline
Mood: Molten
|
|
Nitinol Problems
I've recently been experimenting with nitinol, a shape-memory alloy of approximately equal proportions nickel and titanium.
For those who don't know, nitinol is a shape-memory material in that bending it and then exposing it to heat will cause it to 'unbend', or return to
its original shape. There's plenty of literature about it, but this effectively makes it a powerfully versatile actuator in robotics and elsewhere.
Now, I ordered some nitinol wire from this source to try it out, only to find that it will not activate electrically.
I've extensively tested it, and it works just fine when dropped into a pot of boiling water, or wrapped with kanthal wire. But, for whatever reason,
when hooked up to a circuit, it simply doesn't heat the way I was told it would. Is there something I'm missing about the resistivity of nitinol here?
I have proven that the wire conducts electricity; 2 inches of wire, a 3.7V LiPo battery and a blue LED with resistor confirmed that a circuit was
being made. However, no destressing of the nitinol wire took place even when the voltage was doubled to 7.4 volts.
My guesses are as follows:
-7.4 volts is simply not enough to actuate the wire. Given that the test length was about 2 inches long, this leaves me concerned as to how much
current actually is required.
-The nitinol simply doesn't undergo enough joule heating. Given that a similar material 'Flexinol' does, and is supposed to, I doubt this one.
-The LiPo battery has a current limiter that is keeping the current in the circuit below the required value. I don't have a 'normal' battery on me at
the moment to test this one, but I don't think it's the case - the LED was quite dim even at 7.4 volts.
Does anyone have any suggestions? I'd really like to use this stuff, as I'm working on a robotics project and need 'muscle' actuators.
I would also like to be able to calculate the force one wire can exert as a function of its transition time between the two states (which in turn is a
function of temperature), and very few models seem to exist to help me with this.
Elements Collected:52/87
Latest Acquired: Cl
Next in Line: Nd
|
|
|
Velzee
Hazard to Others
Posts: 381
Registered: 19-8-2015
Location: New York
Member Is Offline
Mood: Taking it easy
|
|
Perhaps these articles are of use to you:
http://www.imagesco.com/articles/nitinol/06.html
http://www.imagesco.com/articles/nitinol/04.html
http://robotics.hobbizine.com/flexinol.html
https://www.physicsforums.com/threads/calculating-force-requ...
P.S. Nitinol ain't too specific XD
Check out the ScienceMadness Wiki: http://www.sciencemadness.org/smwiki/index.php/Main_Page
"All truth passes through three stages. First, it is ridiculed. Second, it is violently opposed. Third, it is accepted as being self-evident."
—Arthur Schopenhauer
"¡Vivá Cristo Rey!"
—Saint José Sánchez del Río |
|
|
Broken Gears
Hazard to Self
Posts: 96
Registered: 7-8-2005
Location: Northern Europe
Member Is Offline
Mood: No Mood
|
|
I'm not sure that the shape-memory is activated by electricity... However, you can control the austenite finish phase formation (AF) temperature very
precise by heat treatment. Most nitinol wire is used in medical devices is the "standard" AF is somewhere around body temperature.
|
|
|
elementcollector1
International Hazard
Posts: 2684
Registered: 28-12-2011
Location: The Known Universe
Member Is Offline
Mood: Molten
|
|
Thanks for the articles, I did learn a few things.
"Nitinol returns to its original shape with a pulling force of about 25,000 pounds per square inch. An actuator wire with a thickness of just 1/100
inch can lift 2 pounds."
I'd presume pulling force scales with thickness (I've confirmed this is true for flexinol, at any rate), so my wire (thickness approximately 0.02
inches, or 1/50") should be able to pull something in the vicinity of 4 pounds (quite a lot for one measly wire...)
That same article also states that for a wire of thickness 0.015 inches (thinner than my wire by a factor of 0.75), the current required is already
2.2 amperes to activate it. With my setup, that would require 3.7 ohms of resistance in the wire assuming the same transition current, which would
occur at a length greater than 18 inches. The current required seems to approximately scale exponentially with respect to wire diameter, though, which
means I might need as much as 4 amperes to activate my wire (occurring at a length of greater than 9 inches).
So, I think this goes a long way towards disproving hypothesis 1 - that the voltage I have is not enough for the wire. If anything, 7.4 volts on just
two inches of wire should actuate it nearly instantaneously. It also disproves hypothesis 3 - that nitinol cannot be electrically driven.
The article does mention that Flexinol and nitinol can be 'destroyed' by pushing too much current through them, though I am unsure as to what this
means. Would they get red-hot? So far, all mine have done is spark slightly when I connect the LiPo battery - which, according to most things I've
read, is merely a normal property of LiPo batteries when being connected. And, again, this nitinol works perfectly fine in all other heating
scenarios.
One thing I did notice during testing is that when I attempted to use a kanthal wire wrapped around the nitinol, it would heat up for a period and
then just... stop. Removing a battery lead and reinserting it restarted the process. So, I think my second guess, that the LiPo battery has some kind
of current limiter, is the correct one so far. If so, it's going to be a doozy to overcome, seeing as I intend those to be my main source of
rechargeable power in the project.
The main difference between the nitinol I bought and Flexinol is that the nitinol I bought does not contract its length, rather, it returns to its
original shape when bent or plastically deformed. I'm not aware of any third type of nitinol, though there might be one that simply doesn't
plastically deform at all (superelastic fishing line, or something).
EDIT: Additional support for the current limiting theory:
https://learn.adafruit.com/li-ion-and-lipoly-batteries/prote...
These batteries have a discharge rate of just 1-2 C?! That's ridiculous! That means that despite their high capacity, they can't even manage circuits
requiring 2-4 amperes. All the other LiPo batteries I've seen have a discharge rate of 25 C, minimum.
(C is a multiplier for how much current a battery can discharge continuously as a function of its charge capacity. For example, a battery rated 2000
mAh (milliamp-hours) with a rating of 2 C can discharge up to 4000 mA (or 4 A) at any given time until it's depleted, which would be about half an
hour at full charge).
I guess that answers my question - I need to find better batteries!
[Edited on 6/9/2017 by elementcollector1]
Elements Collected:52/87
Latest Acquired: Cl
Next in Line: Nd
|
|
|
|
![[*]](images/xpblue/default_icon.gif){kind=icon}
