Idea of "SMA SunMill"

[Khwartz]

Little adding of the second post (in red), to try to make it clearer:

"
Torque in "d' " is:

1 kg * 10 ms^-2 * (d / (pi / 6))

1 kg * 10 ms^-2 * (1 m / (pi / 6))

= ((10 * 6) / pi) Nm =~ 19 Nm.
"

[Khwartz]

NiTi not yet received! 

[Khwartz]

NiTiCu and Nitinol wires received.

NiTiCu is 0.12 mm diameter and Nitinol is 0.50 mm diameter.

Austenite Start temperatures are respectively around 55°C and 80°C.

[Khwartz]

Hi!

After many tries I can say that 0.12 looks useless: to thin too do anything with on a single wire basis, maybe braided or else but not sure.

For the 0.5, I could made some attempts to get "two way memory shapes"; means that after been heat above austenite temperature  (the temperature where the shape corresponds to the denser state), when it cools it comes back at an initial shape, and that is what we want.

Indeed, we want that the "hairs" (if wires around the cylinder), go straight when heated and when they cool recover their initial plied or rolled shape.

I did succeed to make the nitinol 0.5 wire getting straight by just the heat of an incandescent light bulb (but anybody does this) but couldn't obtain more than few percents of self-return to the initial plied shape.

The problem is: "education"!

We are supposed to bring back at cold temperature "a sufficient number of times", the wire after having been straight. I've tried that but only helped with cold water for the cooling, I could get a more important self-return.

Thus, looks to me it is now a matter of having a Shape Memory Alloy allowing us a return to initial shape, and/or a better methodology to "train" the SMA to plie back after having been got straight by the heating phase.

Any help would be appreciated !  ^_^

And sorry, still not have found a way to upload pictures even if less than 5 MO :/

Regards.

[Khwartz]

Still only obtain very few return at home temperature when comes back to martensite (the cold state).

When plied to 90° the wire goes indeed straight (0°) but when it cools back by itself down to room temperature, it self-returns only of 12° of shifting angle.

When cooled by cold water (maximum 10°C I think but of course near 0°C works perfect), we gain 23° more of returning shifting angle.


I have found 3 kinds of shape which might work.

. . . 1. The "O" one, the coil, like curly hairs, but I still need to check it with a larger halogen light so I can better simulate sun rays. Indeed, there is a problem when the coil develops: it should go only up and not touch the cylinder which happens only when only the upper side of the coil is heated but as my heating ray was too much focused I wasn't close enough simulating real situation to check if the development would be really fine and not contradicts with neighbour "hairs" or be stopped by the cylinder surface.

. . . 2. the "S" shape. The big advantage is that you're sure to not have the previous problem. Nevertheless, the difference of gravity centre distance with the cylinder surface looks to me less important (= less "travel" possibility for the wire for the same length of wire). An variant of the "S" shape would be the "8" shape, it is the same but we just "pile" the "S" side to side so they form a "8" and thus, this time, we have a very much closer to the cylinder surface, centre of gravity, like if the "O" shape.

. . . 3. The "L"shape. Just a 90°C angle which looks to have the great advantage on the two previous, to give a faster change shape when heated (the fastest and longest travel with most simple form). The problem with this one is that is we can't obtain apparently more than 23° of auto-return shift angle, it is not very much travel, or we need longer "hairs" indeed so the gravity centre of the wire "travels" more. A variant would be only 20° of wire plying so it goes straight when goes to austenite temperature (the highest temperature changing shape).

I will publish the pictures or even maybe videos on my Google+ "Khwartz".

Regards,
Didier