Kind of motor for joint movement

[Batienzaxcore]

Hey guys. I'm currently a graduating engineering student, and me and my groupmates are on our way to do a portable robot manipulator for knee rehabilitation using a PIC16F877A (We were not allowed to use arduino and arduino clones in our school for this final design project, so yeah, it kinda sucks) and a motor to manipulate the knee using an exoskeletal structure.

But then, I just don't seem to know what kind of motor am I supposed to use, and its specifications such as its maximum torque and supply voltage, since I'm not really good at torque and stuff like this. Any advice what type of motor to use? Thanks.

[Gertlex]

I would imagine the type of motor to use depends on the knee brace or whatever that is being actuated.  More information on that sort of thing would be useful.

[Batienzaxcore]

Would this link help?

http://www.coe.neu.edu/Research/robots/papers/TM_AKROD_ON_LINE.pdf

Basically, the design would be similar to the one illustrated by this project.

[newInRobotics]

I've seen some designs that use out-runner brush-less motor and harmonic drive. I assume stepper motor would do as well, although I'm not sure how much smoothness You'd be able to get with software.

Regarding torque, motor has to have enough torque to lift human body from knee up at a desired velocity and with desired acceleration.

[Batienzaxcore]

Yeah true. Although in most research papers that I've checked regarding this, they don't really tell what kind of motor do they specifically use there 

Another thing is that, the motor would probably have to be operating in a high voltage since it is high torque, right? That's another problem I've been thinking right now, since I didn't want a robot manipulator to be dependent to an ac source, it'll lose its portability.

[newInRobotics]

That's why I've mentioned brush-less motor and harmonic drive.

[jkerns]

Yes, you need a high torque / slow speed output. The paper linked shows a planetary gear reduction system. I would assume that you would buy a motor with a gear reduction already attached. The paper also shows the torque from the brake device - that should tell you about what kind of torque you need to be looking for from your motor /gearbox.

Stepper, DC (brushed or brush-less) - any of those would work with the proper gear box if the combination met your requirements for torque / speed.

The voltage you would need would be whatever the motor you select wants to get.

Obviously, you will have to build / buy a motor controller - either something off the shelf or an H-Bridge.

[Batienzaxcore]

I'm not really sure about the specs since I'm not really that good in English (thank google translate for this), but in checking the specifications of a DC servo motor, I found out that its stall torque is 20kg/cm. So does that mean that it could lift a 20kg object, such as a human leg? If not, then what is the specification that I should look up to in terms of lifting power?

[jkerns]

Torque = force * distance

20 Kg cm would lift a 20 Kg object on the end of a lever 1 cm long. Or it could lift a 2 Kg object on the end of a lever 10 cm long.

The further a mass is from the pivot point, the more torque is required to lift it.

As I mentioned, you will probably need a gear reduction unit - a gear with a 10:1 reduction ratio combined with the 20 Kg cm torque would lift 20 Kg at a distance of 10 cm but the speed would be only 1/10 of the speed without the gears. You multiply the torque at the expense of speed.

How this will come out after going through Google translate is beyond me...

[Batienzaxcore]

Looks like I would have to give up reading Fifty Shades of Grey to understand these kinds of stuff, even if I am an engineering student who is supposed to know this. This would probably be difficult, seeing as the leg of a person varies. I'd have to find a motor with a stronger stall torque. Thanks for the tip I'mma go ask around if ever any more of these problems arises

PS: I was merely kidding about the google translate thing, I believe that my English is enough to communicate with people around the world )

[Batienzaxcore]

Oh wait, I've got some issues here. Torque is force times distance right? Then why is the stall torque labelled as kg/cm? Shouldn't it be kg cm? I multiplied 20 cm on it (Theoretically the distance between the knee and the foot), and got 400kg, which is basically... wrong. I don't think a servo motor could lift about 8 persons with a distance of 20 cm. I'm getting confused here.

That's the link on the servo motor that I saw a few days back. And it says 20kg/cm, not 20kg cm.

http://www.alalso.com/motor/187-20kg-cm-servo-motor.html

[newInRobotics]

With 20cm lever the mentioned motor would stall while lifting 1kg.

[jkerns]

Oh wait, I've got some issues here. Torque is force times distance right?

yes

Then why is the stall torque labelled as kg/cm?

They made a mistake?

Shouldn't it be kg cm?

yes

I multiplied 20 cm on it (Theoretically the distance between the knee and the foot), and got 400kg, which is basically... wrong. I don't think a servo motor could lift about 8 persons with a distance of 20 cm

Bingo.

If it were kg / cm, the longer the arm the more weight you could lift. And that just isn't true.

[Batienzaxcore]

I wouldn't consider it as a mistake though, since I've checked most sites selling motors, and, just the same, almost all of them have stall torques of kg/cm, not kg cm. Or maybe they just used a slash?