Servo or stepper motor for robotic leg?

[NotASmurf]

Hey all, simple question, when building a robotic leg (a small one, bout 30 cm with two partitions, thigh and calf) would one use a stepper motor or a servo? We want the leg to keep its position (as though it were a dc motor attached to a worm gear) and be able to lift 2kg. Any advice appreciated. Also is a separate servo driver required for servos?

 

[Baluncore]

A stepper can trip and miss steps.
Use a DC servo and a position sensor.

 

[NotASmurf]

Position sensor? Are these expensive?
Also will http://www.hobbytronics.co.za/p/377/small-13kgcm-plastic-gears-servo
this be powerful enough to lit 2kg without gear systems?
Does 1.3kg/cm mean it will move a 0.86kg if attached to a 15cm stick? Forgive my begginer-ness

 

[billy_joule]

1.3 kg/cm is a torque value.
It means is can lift 1.3 kg of the end of horizontal 1 cm stick.
or for a 15cm (weightless) stick:
(1.3kg/cm) / 15 cm = 0.087 kg

It could lift 2kg on a 1.3/2 cm stick but that's probably not practical.
Draw a free body diagram of the leg in motion.
You need to consider acceleration (ie how fast you want to move the leg) to get an accurate value for required torque.

Stepper motors will only miss steps if the system is calibrated poorly, 3d printers use steppers and can perform 10 + hour prints without missing a single step.

 

[NotASmurf]

(1.3kg/cm) / 15 cm = 0.087 kg
(kg/cm)/cm = kg
kg/cm^-2 = kg

So surely it should be kg/cm*cm =kg? ie 1.3 kg/cm *15cm = 19.5??

 

[billy_joule]

Oops. The problem is you used the wrong units for Torque, I ignored that and used it as a torque value without correcting the units.
It should be kg*cm not kg/cm so 1.3 kg*cm

T = Fr
Torque (Nm) = Force (N) X Radius (m)

or in your non SI units:
kg*cm = kg X cm
(using weight instead of force)

A lesson in always checking what the units represent physically for you - Your units are consistent but don't make sense.
A lesson in not copying and pasting values without checking units for me!

mass/distance represents linear density eg a rope or wire has a mass/length (maybe 0.02 kg/m for an average rope? I don't own a rope to check)
Mass/distance makes no sense as a motor output parameter.

 

[Mike_In_Plano]

Brushed core-less and brush-less motors give the most bang per lb when you consider power and efficiency. Maxon motor and micro motor produce awesome core-less motors and can offer them with planetary gears.

 

[NotASmurf]

can offer them with planetary gears.

Is there a name for them when they have internal planetary gears to up the power? I need to know what to search for, I'm in south africa so there arn't many places I can go to :P. there is a local site that sells though hobbytronics.co.za

 

[meBigGuy]

Stepper motor gives you precise position relative to steps with little overshoot or damping. I've used them in wafer probers and large xray emitter positioning. A servo requires feedback and response tuning and might oscillate a bit about the final position. A stepper can slip if given too large a load, or a mechanical shock. If so, it won't recover, whereas a position based servo will recover.

A stepper operated open loop (no position sensing) need to be calibrated somehow.

I've never tried to compare torque, etc, but I'd expect a servo to be more energy efficient and potentially faster. I'd expect steppers to be easier to implement.