walker robot

[TeslaHV9]

Hi-
 I was thinking about making a small, humonoid walking robot. It would use thin nitinol wire (a.k.a muscle wire) for the leg contraction. Althought the nitinol would need a high ampage source about the size of a car battery to make it shrink a substatial amount (that would not fit in the planned size of the robot). I had the idea to wrap the nitinol wire in nichrome wire (burns or heats up when eletricity is passed through it) to make the nitinol shrink because only little voltage and ampage is required. I am wodering if anyone has any suggestions to this or suggestions on the topic of the robot's legs. thanks 

[paulstreats]

ive seen muscle wire work with a single 9volt battery like you put in a walkie talkie.

 You really just need different lengths for a range of movements, a single piece of muscle wire is either 1 extreme or the other so you would need say 20 pieces per muscle to allow for a 20 step motion resolution.

[Soeren]

Hi,

nitinol would need a high ampage source about the size of a car battery to make it shrink a substatial amount (that would not fit in the planned size of the robot). I had the idea to wrap the nitinol wire in nichrome wire (burns or heats up when eletricity is passed through it) to make the nitinol shrink because only little voltage and ampage is required.

You only need high amps if you use very wide NiTiNol wires.

The power needed to heat up your wire will be the same no matter the method!
Using a wrap around of NiChr only means that you will loose more power to the exterior, since it will have a larger part of the surface in the air, rather than toward the NiTiNol, so it will use more power and will react slower - both ways, heating and cooling.

Have you got a datasheet (or a link to it) for your wire?

[Soeren]

Hi,

You really just need different lengths for a range of movements, a single piece of muscle wire is either 1 extreme or the other so you would need say 20 pieces per muscle to allow for a 20 step motion resolution.

One piece and PWM control (just like used in a cars gas injectors) should do, if the wire used is thin (to cool quickly).

[dellagd]

couldn't you just have this:
one strip of Nitonal (or however many it takes ) and run it through the leg like this:
PS the little right angles are to stop the leg from bending to far.
the springs are for it to snap back into place when the nitonal stops its pull

[Razor Concepts]

Nitinol is a very wasteful... only really useful for micro-sized robots. I guess you could use it on bigger robots but it won't have any advantages over a robot with servos on the legs.

[dellagd]

what about cost?

[TeslaHV9]

thanks for your help. I will see if I can get the nitinol to shrink with a 9v. i guess i will abandon the idea to use nichrome. if you have any more suggestions about making a walking robot please tell me. also the robot is planed to be very small, about the size of a deck of cards. i would like to attempt to use nitinol before going to servos because nitinol is alot cheaper from where i can get it.
thanks
 TeslaHV9

[SmAsH]

wow, thats one small walker. this will truly be a challenge! i wish you good luck on this endeavor! but what mcu will you be using that is small enough?

[dellagd]

I know I was asking smash and others who posted on your topic to see what they would say about the efficiency if the cost was very low
PS plz don't mention my name to the whole world and edit it out
edit:
thanks tesla

[Ro-Bot-X]

You might get some ideas from this product:
http://www.unitedhobbies.com/UNITEDHOBBIES/store/uh_viewItem.asp?idProduct=6405&Product_Name=Toki_BioWire_1g_Micro_Servo_(world_first)

or this one:

http://www.microflight.com/Online-Catalog/Actuators-and-Servos/Musclewire-Servo-RC-1

[TeslaHV9]

those servos cost about $30 each! I can get the nitinol for around $3-5 for plain wires