# Controlling flexinol with an arduino board?

I am looking to wire up 16 pieces of flexinol (http://www.musclewires.com/MWFlexinol.php) using an arduino board. Ideally I would like to be able to activate them independently, meaning that they could all be on, off or in any combination in between. My understanding that they would need to draw more current than the board could handle so some type of shield would be necessary?

Any suggestions would be greatly appreciated.

First thing's first, this just seems to be a wire that has special characteristics. If so, we just need to find out the resistance and use Ohm's law.

From the product page, I found this image:

I can tell from that that it has a resistance of 20 ohms and a recommended current of 1 amp. First, let's see how much current it would draw with a 5V supply. I prefer using a Ohm's law triangle, just because I'm a very visual person. So with this math:

5 / 20 = 0.25

That's 250 mA, but an Arduino can only provide 40 mA! Furthermore, it is only 1/4 of the recommended current that the manufacturer specifies.

To get the full current the OEM specifies you'll need a 20 V power supply capable of at least 1A. Even then, you might want to put in a power resistor to limit the current.

Driving this with Arduino: So, we have a few problems here:

• It is 20W, which is quite a bit in the Arduino world
• We need to control the current/voltage dynamically
• You need 16 of them, so we don't want something too costly
• You probably can't find something to fit your needs that is one shield.

What comes to mind is a transistor. I'm not skilled with picking them out, but there are a lot of guides online. You'd then want to add a capacitor and a resistor to make some sort of filter and connect it to a PWM pin to send a varied voltage to the transistor, which will vary the amount of power going through it.

Alternatively, you could use a relay. However, that only does ON or OFF. A solid state relay (SSR) would allow you to use PWM on the relay and then you could add a capacitor/resistor combo on the other end of it, but you'd need a pretty large capacitor or a pretty large frequency for the PWM (which you can modify).

• "You'd then want to add a capacitor and a resistor to make some sort of filter and connect it to a PWM pin to send a varied voltage to the transistor" - NO. This is both unnecessary and highly inefficient. The thermal mass of the wire will average out the PWM quite well, and it is far more efficient to use a transistor (or better FET) switch which is (to the greatest degree possible) only ever all the way on or all the way off, so keep the control signal to the transistor digital and do not average it with a filter which would cause it to spend time in the lossy "somewhat on" region. Sep 30, 2014 at 20:06